Courses & Programmes

Accounting is the language required to do business. It allows executives, managers and employees to communicate; it allows investors, analysts, and governmental authorities to understand how an organization is doing. Thanks to the International Financial Reporting Standards, accounting can be seen as a global language, spoken across most countries.

Accounting is much more than book-keeping. Without accounting, it is not possible to evaluate a new business opportunity, assess the performance of a company, or design a solid business strategy for the near future. This course provides a thorough introduction to financial and managerial accounting. It provides fundamental knowledge that all business practices require.

This course teaches how to formally record and report economic events and transactions, how to read accounting information to make inferences and support decisions, and how accounting plays an active role in the success or failure of a company. More broadly, this courses explains why the accounting activity cannot be performed by robots, as it involves discretion in how information is recorded and reported. Moreover, the course offers a first look at what the accounting profession entails.

Wil jij leren hoe je een bedrijf kunt beveiligen tegen een cyberaanval? Vind je het leuk om te ontdekken hoe de beveiliging van computersystemen in elkaar zit? En ben je op zoek naar een praktijkgerichte opleiding, maar wil je geen vierjarige hbo-bachelor volgen? Dan is de tweejarige Associate degree Cyber Security misschien iets voor jou!

This course covers both theoretical and practical aspects of complex
dynamic econometric models that are used in the industry, by central
banks, governments, think tanks, and other research institutes.

The students will be introduced to stochastic theory that allows them to
fully understand the dynamic properties of complex models featuring
nonlinearities, time-varying parameters and latent variables. Important
concepts include invertibility, stationarity, dependence, ergodicity and
bounded moments.

The students will also be introduced to advanced estimation theory that
allows them to bring state-of-the-art models to the data and conduct
inference on parameters under very general conditions. Important topics
include the existence, measurability, consistency and asymptotic
normality of extremum, M and Z estimators. We also cover advanced topics
in nonlinear model selection and specification, estimation and inference
under incorrect specification, metric selection, structure and causality.

From a practical perspective, the advanced methods and state-of-the-art
models are used for forecasting and policy analysis in a wide number of
applications ranging from finance to macroeconomics and data science.

The aim of the Advanced Econometrics 1 course is to obtain a deep understanding of econometric theory, practice and inference using a variety of advanced econometric techniques.

After passing the course, students should be able to apply advanced econometric techniques in practice, to extend currently available methods when needed for particular applications, to implement these methods in a matrix programming environment, and to understand and derive their statistical properties.

The aim of the Advanced Econometrics 2 course is to obtain a understanding of several advanced statistical techniques, to conduct  and evaluate inference and report about statistical analysis in a critical way. These techniques will be implemented using software (like MATLAB, R or Python). The contents of this course build upon the general knowledge acquired in the course Advanced Econometrics 1.

Topics include: bootstrap methods; semi- and non-parametric methods; weak identification; panel data models.

This course is part of the Mastermath programme. All information can be found at the Mastermath website.

Machine learning is the science of getting computers to act without being explicitly programmed. Machine learning is so pervasive today that it is used in everyday life without knowing it. In this course, you will learn about the most effective machine learning techniques, and gain practice implementing them and getting them to work yourself. We will discuss the theoretical underpinnings as well as the practical know-how needed to apply these techniques to new problems.​

The course covers recent advancements in networking technologies in
general-purpose computing (e.g., data center, cloud computing). The
course is split into two themes: End-host networking and network
infrastructure.

In the end-host networking theme, students will learn about networking
concepts (MTU, Segmentation, stateful and stateless Offloading, RSS,
SGE, multiqueue), internals of the Linux networking stack, challenges
associated with delivering high-performance network I/O operations (100+
Gbps bandwidth, millions of network operations/sec, many/multi-core
scalability), alternate ways of building networking stacks in the
userspace, design of packet processing stacks, and introduction to a
high-performance Remote Direct Memory Access (RDMA) technology.

The second theme of the course will be focused on the recent development
in network infrastructure. Students will learn about the architecture of
the network inside a large-scale data center of companies like Amazon,
Google, and Microsoft, new ways of architecting a computer network
(software-defined networking), technologies for improved network
programmability (software switches/routers, network virtualization,
programmable data plane), and how to utilize these new technologies to
build new applications/services.

Please refer to the System and Network Engineering web pages for detailed and current course information.

The course will feature a number of hands-on assignments accompanied by lectures on advanced operating system kernel design and programming concepts. In each assignment, students will be expected to start with a minimal kernel implementation and exercise their kernel hacking skills on one of the major operating subsystems (i.e., memory management, process management, drivers, etc.). This will involve programming in both C and assembly as well as directly interfacing with the hardware. The course will also link lectures and assignments to modern operating system features and offer insights into state-of-the-art OS research efforts.

To learn advanced programming skills, to get to know and understand advanced programming concepts like inheritance and to get experience with programming some of the data structures that were taught in the course Data Structures & Algorithms.

The student will have an appreciation of modern programming paradigms as realised in a range of current programming languages.
The student will have achieved an understanding of how programming languages are interpreted and compiled through the practical construction of interpreters.
The student will be able to quickly assess and acquire a novel programming language.

Learning Outcomes
Students who have completed the course will have an understanding of the principles of programming languages and the methodological foundations of computer programming as well as the demonstrated ability to apply these principles. They will have highly-developed analysis and problem solving skills as well as the ability to quickly acquire and program in a range of programming languages.

In this course we will cover a series of techniques that are more advanced than those covered in BRMS1, BRMS2, or Statistics for Sciences. We will work extensively with science and social science data and learn how to analyze and interpret data at an advanced level. The course covers the following topics:

– Review of multivariate linear regression and ANOVA
– Basic matrix algebra
– Model selection and diagnostics in multivariate linear regression
– Logistic regression
– (M)ANOVA and (M)ANCOVA
– Discriminant analysis and classification
– Principle component analysis
– Analysis of repeated measures
– Categorical data analysis and multi-way frequency tables

The central aim of the course will be that students acquire the skills to conduct and interpret quantitative analyses of various empirical studies. Students will also analyze science and social science data using the techniques learned in the course and complete data analysis assignments.

Please refer to the System and Network Engineering web pages for detailed and current course information.

At the end of this module a student is able to:
• correctly reproduce the central concepts of multiple regression analysis;
• independently and correctly calculate statistics such as the (un)standardized b-coefficient, intercept, Sum or Squared Errors, Total Sum or Squares r, R², f and t values, and standard deviations;
• independently and correctly perform multiple regression analysis with interaction-effects and categorical independent variables with statistical software;
• independently and correctly interpret the results of statistical analysis performed by themselves and/or by statistical software;
• independently and in a correct, precise and clear way write up results of quantitative hypothesis-testing research regarding variability among different social groups.

This course provides a first introduction to advanced topics in data
analysis that you will likely encounter when working in the field of
health sciences. We will explore the identification and handling of
clustered data that challenge the assumptions of standard statistical
techniques. We will also focus on the analysis of longitudinal data that
are seen frequently in the field of health sciences. Finally, we will
touch on the estimation of direct and indirect effects which can help us
to disentangle results from complex interventions.

Although the course expands knowledge of, and hands-on experience in
analytical techniques, it has not the aim to make you an expert in these
topics. This course does provide you with the confidence to engage in
discussions surrounding complex analytical approaches in the field of
health sciences.

This course provides an introduction to mathematical modeling of computational problems. It covers the common algorithms, algorithmic paradigms and data structures used to solve these problems. It also uses the Python programming language to implement and test algorithms and data structures on realistic datasets. The technological topics which will be covered in this course are:

• Python Programming Basics;
• Introduction to Object-Oriented Programming in Python;
• Algorithm Analysis;
• Basic Data Structures:
• Recursion;
• Sorting and Searching;
• Trees and Tree Algorithms;
• Graphs and Graph Algorithms.

Have you ever wondered how we can track a gene across 3 billion years of
evolution? Or how you can use the genome information of a given cancer
patient to find out what may be wrong? Sequence alignment can be used to
compare genomes, genes or proteins from bacteria all the way to humans,
while further derived algorithms may be employed to make a phylogeny (to
find out about evolutionary relationships), find a functional motif in a
protein sequence, or a viral sequence in a genome. In this course we
focus on the most important algorithms for biological sequence analysis
that can be applied to real scientific problems in biology.

After this course students will be able to:

– Develop insights into personal strengths, weaknesses, core values and development priorities;

– Develop the ability to inquire and advocate in an effective way;

– Understand and apply different styles of influencing with integrity;

– Reflect on the effectiveness of one’s leadership behaviours by applying practical concepts;

– Develop insights into importance of diversity and inclusiveness in leadership;

– Create a culture of learning and giving/receiving high quality feedback;

– Generate an effective team charter in order to maximise the impact of team work.

In this 4-day interactive course you will learn key skills to perform as an Analytics Translator. You will play the bridging role between the technical expertise of data scientists and the operational expertise of domains such as marketing, HR, supply chain, finance. This role is crucial to ensure that the data science efforts connect flawlessly to the business needs. This unique course will prepare you for this new and valuable role in every organisation. The next edition will start on Thursday 2 September 2021.

This course gives an introduction to financial mathematics.
The following subjects will be treated:
– introduction in the theory of options;
– the binomial method;
– introduction to partial differential equations;
– the heat equation;
– the Black-Scholes formula and applications;
– introduction to numerical methods, approximating the price of an
(American) option.

This course is about regression analysis for causal inference and prediction, which is a powerful tool used in empirical economics and finance. In the part on causal inference, estimation and testing of the basic linear regression model by ordinary least squares (OLS) will be reviewed. Particular attention will be paid to common failures of the exogeneity assumption and how these can be solved using an appropriate empirical design. These include experimental-, instrumental variable-, regression discontinuity-, and fixed effects-designs, and event studies. In the part on prediction, attention will be paid to developing econometric models for forecasting.

The objective of this course is to give Business students the necessary econometric tools to complete te Finance specialisation; Finance is a field with strong emphasis on empirical work and rich datasets. Upon successful completion of this course students should have the following knowledge, skills and attitudes:

• a solid knowledge of the various types of datasets used in Empirical Finance, and their corresponding techniques (time-series, cross-sectional, event-study and panels);
• a solid understanding of the assumptions underlying regression analysis;
• a solid knowledge of the statistical tests and techniques used to detect and remedy violations of these assumptions;
• a solid understanding of the factors influencing the significance found in empirical research;
• be able to conduct statistical tests on regression results in order to answer research questions.

All of this will have a strong focus on applied skills.

Public policies need to be evaluated in order to understand their
effectiveness and correct validation of economic theory can only be
achieved with empirical research. The main objective of this course is
to provide an overview of econometric research methods in urban, real
estate, transport and environmental economics and to teach you how to
apply these methods to real-world data. After following this course, you
will:
• have an advanced understanding of the mathematical and statistical
concepts underlying regression analyses in spatial economics;
• understand the importance of and difficulties in estimating causal
effects as opposed to correlations in spatial economic problems;
• know how to appropriately interpret regression results of various
estimators and know which one to apply in particular situations,
depending on (i) the nature of the data (cross-sectional / panel /
discrete data) and (ii) the task at hand (i.e., valuation of public
policies, testing of economic theories or estimating parameters as
derived from theory);
• understand and know how to apply techniques that are commonly in use
in urban, real estate, transport and environmental economics and policy:
spatial econometrics, discrete choice models and quasi-experimental
research designs;
• be able to apply these methods independently to typical datasets in
spatial economics using the software package STATA.

Public policies need to be evaluated in order to understand their effectiveness and correct validation of economic theory can only be achieved with empirical research. The main objective of this course is to provide an overview of econometric research methods in urban, transport and environmental economics and to teach you how to apply these methods to real-world data. After following this course, you will:
• have an advanced understanding of the mathematical and statistical concepts underlying regression analyses in spatial economics;
• understand the importance of and difficulties in estimating causal effects as opposed to correlations in spatial economics problems;
• know how to appropriately interpret regression results of various estimators and know which one to apply in particular situations, depending on (i) the nature of the data (cross-sectional / panel /discrete data) and (ii) the task at hand (i.e., valuation of public policies, testing of economic theories or estimating parameters as derived from theory);
• understand and know how to apply techniques that are commonly in use in urban, transport and environmental economics and policy: spatialeconometrics, discrete choice models and quasi-experimental research designs;
• be able to apply these methods independently to typical datasets in spatial economics using the software package STATA.

This course is about regression analysis for causal inference and prediction, which is a powerful tool used in empirical economics and finance. In the part on causal inference, estimation and testing of the basic linear regression model by ordinary least squares (OLS) will be reviewed. Particular attention will be paid to common failures of the exogeneity assumption and how these can be solved using an appropriate empirical design. These include experimental-, instrumental variable-, regression discontinuity-, and fixed effects-designs, and event studies. In the part on prediction, attention will be paid to developing econometric models for forecasting.

Machine learning is marking a revolution in the world. From an academic research topic, over the last decade it has shift to a major paradigm used in many companies for a wide range of services. From deleting SPAM mail from your inbox to ranking the Google search results, and from defining your Facebook stream to enabling medical diagnoses.

In this course we study learning from large collections of unstructured data, such as text documents, web pages, images and videos. We address the complete machine learning chain, from designing the system and its objectives, to representing data and selecting and evaluating the learning method. We review and focus on the foundations of supervised classification and regression, which we extend to deep learning. We have application lectures on text processing, computer vision and recommender systems.

You will learn the theoretical concepts during the lectures with a keen eye on the design of the full learning system. In the tutorials we will focus on some off the important mathematical concepts, and in the lab you will gain hands-on experience through a number of coding assignments (implementing your own deep learning functions) and by participating in a Kaggle competition or project. Finally, a few experts from the field (both academic as well as industry colleagues) are invited to provide guest lectures.

This course introduces the basic concepts underlying applied statistics.
Its main focus is on estimation of effect sizes and confidence
intervals, and elementary statistical tests. The applied techniques that
will be introduced are aimed at description of observed data, and
estimation of and testing null-hypotheses about single population means
and proportions, differences between means and proportions, contrast
analysis of more than two means in independent and repeated measures
designs, and correlation. Tests that will be introduced are the t-test,
chi-square test, and the ANOVA F-test.

This course deals with a number of stochastic modeling techniques that are often used in practice. They are motivated by showing the business context in which they are used. Topics we deal with are: time-dependent Poisson processes and infinite-server queues, renewal processes and simulation, birth-death-processes, basic queueing models, and inventory models. We also repeat and extend certain parts of probability theory.

The development of modern computer technology requires professionals with a background in all science fields, who understand both hardware and software. The interaction between the hardware and software on a variety of levels provides a framework for understanding the fundamentals of computing. Whether your primary interest is hardware or software, computer science or electrical engineering, the central ideas remain the same. This course will show the relationship between hardware and software, and focus on the concepts that are the basis for today’s computers.

This course is only available in Dutch.

The computational turn has fundamentally changed the way people and institutions produce, use, collect, share, and store information. Data is the vital fuel for a growing number of applications in daily life and has become a fundamental human resource, similar to food, energy or transportation. The result is a fragmentation and personalisation of data creation and data use.

What are the implications of these developments for information management and recordkeeping? How can we make sure relevant information remains accessible throughout time? How does our data-driven culture transform societal values such as transparency and privacy? And how do we keep the archives of the past connected to the needs of the present? How do we design more inclusive archival infrastructures which give a voice to underrepresented groups in the past and the present? These questions are central to the Dual Master’s in Archival and Information Studies.

Artificial Intelligence (AI) is widely used in our society: from cars that detect pedestrians to our smart phones’ virtual assistants. These applications use AI techniques to interpret information from a wide variety of sources, and in turn to enable intelligent, goal-directed behaviour.

The Artificial Intelligence Master’s programme at Vrije Universiteit Amsterdam looks specifically at hybrid intelligence, where AI systems and humans collaborate. The first year is made up of broad courses that focus on core AI topics, while the second year is devoted to your chosen area of specialisation.

In collaboration with Psychology department, we also offer the specialised Cognitive Science track.

AI for Health is a new track in collaboration with Medical Informatics (Amsterdam UMC, UvA) that will start in the academic year 2021-2022. In this track you’ll learn about Medical Informatics basics, Medical AI (with a focus on imaging techniques in medicine and natural language processing techniques in medicine), and how to combine machine learning and reasoning for health applications. Two courses will take place at Amsterdam UMC, location AMC.

As an Artificial Intelligence student, you study and analyse intelligence intuitively. For example, you research computers to imitate human thinking and acting in order to perform better. Human intelligence and the relationship between brain and behaviour is an important source. In addition to mathematics, computer science and programming, you will therefore take courses in cognitive psychology, logic, linguistics and philosophy. With this knowledge you learn innovative, learning computers and develop, develop and build.

Do you often think about how everyday actions can be made using computer technology and algorithms? Do you like mathematics and do you also have a feeling for language and logic? Then the study Artificial Intelligence at the UvA might be something for you.

This programme is only available in Dutch.

This course is part of the Mastermath programme. All information can be found at the Mastermath website.

This course is designed to provide students with the background in discrete probability theory and programming that is necessary to follow other more advanced master-level courses in areas such as linguistics, natural language processing, machine learning, complexity theory, cryptography, information theory, quantum computing, combinatorics, etc. The goal is to make students that have had no prior exposure to probability theory and/or programming feel comfortable in these areas. To achieve this goal we will try to illustrate the theoretical concepts with real-life examples that relate to topics in, e.g., computer science, gambling, and the like. Moreover, we will make sure that there is a close tie between the theoretical and practical part of the course, thus enabling students to apply their newly acquired theoretical knowledge to real problems.

This course is designed to provide students with the background in discrete probability theory and programming that is necessary to follow other more advanced master-level courses in areas such as linguistics, natural language processing, machine learning, complexity theory, cryptography, information theory, quantum computing, combinatorics, etc. The goal is to make students that have had no prior exposure to probability theory and/or programming feel comfortable in these areas. To achieve this goal we will try to illustrate the theoretical concepts with real-life examples that relate to topics in, e.g., computer science, gambling, and the like. Moreover, we will make sure that there is a close tie between the theoretical and practical part of the course, thus enabling students to apply their newly acquired theoretical knowledge to real problems.

Social science is best characterized as an ongoing activity, namely the creation of new knowledge. This activity follows rather strict rules in order to gain acceptance by the scientific community.

This course covers all discipline-independent aspects of creating new knowledge:

• How to formulate a scientific question

• How to plan an investigation bearing on that question

• How to conduct the inquiry

• How to present the data that result from your research

• How to interpret your results, and extrapolate beyond your data

• How to report the results in an appropriate way

The central question we will address in this course is this: How do I conduct research such that it will yield conclusions that are acceptable to critical peers? Central concepts:

• The Empirical Cycle: start with a problem or question, conduct experiments / observe reality, interpret data in light of the original problem, form a tentative conclusion, make predictions.

• Experimental design: organize experiments / observations such that they allow unequivocal interpretations.

• Statistics: a branch of probability calculus that deals with data analysis. It aims to extrapolate the results of an investigation beyond the boundary of the sample by adding an element from outside the investigation: a mathematical description of the data. This addition allows for making quantitative judgments about the meaningfulness of the data.

• Communicating results: presenting findings using oral and written reports.

The central aim of the course will be that students acquire the skills and knowledge to conduct research, and quantitatively analyze data. The central question we will address in this course is: How do I design research and analyze data such that it will yield conclusions that are acceptable to critical peers? On one hand, this course will further students’ quantitative skills developed in BRMS I (1st year course); various quantitative methods commonly used in social science research are covered.

On the other hand the steps in designing research (formulating research questions, operationalizing, planning and conducting an empirical study) will be treated.

The course will consist of an alternating series of interactive lectures and practicals in which students learn the theoretical background as well as the application of several statistical techniques, including: correlation, reliability, multiple linear regression, and ANOVA. The practicals will mostly be used to learn coding, analyzing, reporting and interpreting data using SPSS.

In order to become a skilled and versatile methodologist, you will need a rich methodological toolbox. But before acquiring those tools, in this first course of the first semester of the PML specialization, you will build the box itself. Boring you say? No! The box will give you the fundament for all the advanced tools you will learn during the remainder of the programme. You will build your box from the basics of mathematics, statistical programming, and statistics. The more solid these fundaments are, the fancier your box will be, and the better you’ll be able to build your statistical toolset during the subsequent courses.

Mathematics

In this module, you will improve your algebra skills. You will simplify and solve equations involving logarithmic and exponential functions. In addition, you will learn how to differentiate functions. These skills will help you to compute maxima and minima of functions. Finally, you will learn how to conduct matrix algebra.

Statistical programming

In this module, you will learn the programming language R (and thereby fully replace SPSS). We will focus on working with datasets, using and programming your own functions, plotting and basic programming routines (if-statements and loops).

Frequentist statistics

In this module, you will use R simulations to gain a conceptual and practical understanding of fundamental concepts in statistics. We will cover elementary probability theory (e.g., marginal, joint, and conditional probabilities), probability distributions (e.g., discrete and continuous distributions), hypothesis testing (e.g., central limit theorem, standard deviations, p-values, confidence intervals, power), and regression analysis (e.g., polynomial regression, model selection, overfitting).

This course (named Computational Econometrics (E_EOR3_CE) in the past
academic years) will cover Bayesian statistics where the topics include
the prior and posterior density, Bayesian hypothesis testing, Bayesian
prediction, and Bayesian Model Averaging for forecast combination.
Several models will be considered, including the Bernoulli/binomial
distribution, the Poisson distribution and the normal distribution.
Obviously, attention will be paid to the Bayesian analysis of linear
regression models. Also simple time series models will be considered. An
important part of the courses is the treatment of simulation-based
methods such as Markov chain Monte Carlo (Gibbs sampling, data
augmentation, Metropolis-Hastings method) and Importance Sampling, that
are often needed to compute Bayesian estimates and predictions and to
perform Bayesian tests.

This course will cover Bayesian statistics where the topics include the prior and posterior density, Bayesian hypothesis testing, Bayesian prediction, and Bayesian Model Averaging for forecast combination. Several models will be considered, including the Bernoulli/binomial distribution, the Poisson distribution and the normal distribution. Obviously, attention will be paid to the Bayesian analysis of linear regression models. Also simple time series models will be considered. An important part of the courses is the treatment of simulation-based methods such as Markov chain Monte Carlo (Gibbs sampling, data augmentation, Metropolis-Hastings method) and Importance Sampling, that are often needed to compute Bayesian estimates and predictions and to
perform Bayesian tests.

In frequentist statistics we assume that the data is distributed according to some unknown probability distribution. In Bayesian statistics, the data and the parameter are both treated as a random variable. Besides specifying the statistical model, the Bayesian procedure also specifies a prior distribution on the model. The data will be used as an updating mechanism for the prior resulting in the posterior distribution.

In this course we consider the consider the classical problems considering point-estimation, hypothesis testing, confidence sets and decision theory where we will describe the Bayesian and frequentist methods and compare them to each other. Furthermore, we will discuss the choice of the prior distribution, depending on both the statistical model and the intended posterior distribution.

This course is part of the Mastermath programme. All information can be found at the Mastermath website.

At the end of this course, students can (a) name key components of Bayesian statistics (paraphrasing), (b) paraphrase basic aspects of Bayesian statistical methods, and (c) analyze psychology papers that use these methods. Based on an evaluation of the studied materials, students are also able to (d) program rudimentary probabilistic models and statistical methods (scientific thinking) that are then used to (e) analyze practical research questions (evaluation). Students can (f) report their thinking in short essays (written communication and reflection).

Modern machine learning methods are based on mathematical concepts, especially from probability theory and statistics. This course treats these concepts in detail, through the spectrum of the Bayesian school of thought in machine learning. This will lay the groundwork for a solid understanding of advanced machine learning methods taught in other courses.

This course is only available in Dutch.

The course focuses on the modeling, analysis and optimization of
complex decision making processes which involve human behavior (such as
selfish, risk-averse, altruistic or malicious behavior). Building on
game-theoretic foundations, you learn to model processes of complex
decision making and to quantify the inefficiency caused by human
behavior.

The main goal of the course is to equip you with algorithmic
optimization techniques to master the challenging task of reducing the
inefficiency of such processes. These techniques find their applications
for example in Traffic Routing, Network Design, Cost Sharing,
Resource Allocation and Auction Design.

Understanding data about human behaviour is an important and valuable skill in today’s society. The police use data to predict the risk of burglary by area and week of the year, whereas insurance companies adjust their prices based on client data, and schools adjust educational programmes based on what is known about student progress. Companies, public institutions and governmental organisations: they all use the continuous stream of big data to describe and predict human behaviour.

This course will provide students with a general understanding of data-related and systems-related challenges in Big Data applications. They will gain fundamental knowledge about principled approaches to tackle such challenges, with respect to systems abstractions, programming models and execution models for parallel and distributed data-intensive applications.

The course prepares students  for data-related tasks in a job as Data Engineer, ML Engineer, Applied Scientist or Researcher, and puts a focus on their implementation skills using the programming language Java, for example by walking students through low-level MapReduce jobs for several data related problems and common data processing operators. At the same time, the course highlights ongoing research problems in the area of Big Data processing. Furthermore, the course details the history of many systems currently at the forefront of computing, e.g., it discusses the roots of Google Tensorflow in previous Big Data systems at Google.

The course will also feature guest speakers from various companies to connect students to real world problems.

“Big data” is a relatively new phenomenon, and refers to data that are more voluminous, but often also more unstructured and dynamic, than traditionally the case. In Communication Science and the Social Sciences more broadly, this in particular concerns research that draws on Internet-based data sources such as social media, large digital archives, and public comments to news and products. This emerging field of studies is also called Computational Social Science (Lazer et al., 2009) or even Computational Communication Science (Shah, Cappella, & Neuman, 2015).

The course will provide insights in the concepts, challenges and opportunities associated with data so large that traditional research methods (like manual coding) cannot be applied any more and traditional inferential statistics start to loose their meaning. Participants are introduced to strategies and techniques for capturing and analyzing digital data in communication contexts. We will focus on (a) data harvesting, storage, and preprocessing and (b) computer-aided content analysis, including natural language processing (NLP) and computational social science approaches. In particular, we will use advanced machine learning approaches and models like word embeddings.

Big data analytics encompasses machine learning techniques for extracting insights from large amounts of data and building predictive models. After this course students will be able to a) summarise various key concepts in big data analytics (paraphrasing), and will have hands-on experience with these techniques applied to a wide range applications using R (analysing, evaluating); b) handle unstructured data from various sources; use the grammar of data manipulation to slice and dice data with tidyverse; build predictive and analytic machine learning models and evaluate their performance; c) conceptually explain a wide range of prominent machine learning techniques (paraphrasing, communicating) and understand their relation to statistical and psychometric methods learned in bachelor courses (evaluating, reflecting); d) extract predictive features from various types of unstructered data-sources (text, signals, images, etc.).

Machine learning methods covered: Supervised: regularized regression, trees, bagging, boosting, random forests, k-nearest neighbours, logistic regression, discriminant analysis, support vector machines, (convolutional) neural networks; Unsupervised: PCA, clustering, k-means.

This elective addresses important concepts in bio- and neuroinformatics
and big data analysis, with powerful applications in neurosciences. Lectures
and practical assignments provide theory and hands-on experience in fast
moving fields of exploratory, graph and predictive data analytics,
neuroscience, connectomics and metagenomics.

Due to the growing digitisation in society, increased
international collaboration, data sharing and technological advances,
there is a vast increase in the volume and complexity of data sets that
are analysed in modern science. One social media experiment, brain
imaging study or DNA sequencing project easily encompasses terabytes of
information.
Large data volumes, commonly referred to as “Big data”, have many
advantages (more information, increased statistical power, mining for
previously unknown relations) but come with the need for special
strategies and approaches both to manage the data (transfer, storage,
updating and sharing) and process and analyse the data (cleaning,
visualization, querying, statistics, mining). In addition concepts such
as ethics and information privacy need to be considered. In this course
large data sets from experiments in cognitive and/or biological
psychology are the central theme. In the individual tutor meetings the
students obtain insight in the following steps: obtaining and examining
datasets and drawing-up hypotheses based on classical methods of
cognitive / biological psychology; preparing the data for analysis;
performing the analysis and interpreting the outcome. The students write
their own research proposal and present the proposal using PowerPoint.

In this module we dive into cloud technologies that allow organizations to tap into potentially thousands of computers at the click of a button at little upfront cost. We also explain the software that is used to do this and also to program such compute clusters, in order to use them for addressing Big Data problems. More info here: www.cwi.nl/~boncz/bads

Please note that this course cannot be followed separately.

Nowadays it is easy to measure, collect and store a large number of
observations. However, many of the traditional inferential methods were
developed for small data sets. While the questions traditional
inferential methods tried to answer like ‘What can we learn from the
data’ or ‘How can we use the data to make predictions for future
observations’ are as important as they were many years ago, the methods
we need to answer these questions when confronted with big data sets
have been developed only recently. In this course, we will first study
the methods that have been introduced for hypothesis testing of large
data sets. Next, we learn more about regression models for big data
sets. Moreover, we discover some of the possible pitfalls arising from
electronic computation and study selective inference.

In this course we dive into cloud technologies that allow organizations to tap into potentially thousands of computers at the click of a button at little upfront cost. We also explain the software that is used to do this and also to program such compute clusters, in order to use them for addressing Big Data problems.

Deze cursus heeft als belangrijkste doel het praktisch te leren omgaan
met kleine en grote biologische data sets. Technieken voor het werken
met bestanden en databases, het samenvoegen van tabellen, het grafisch
weergeven en analyseren van grote data sets en  reproduceerbaar werken
zullen behandeld worden. De technieken worden geoefend aan de hand van
gegevens uit verschillende vakgebieden van de biologie. De gebruikte
programmeertaal is R.

This course is only available in Dutch

This course addresses the relationship between human rights and the use
of big data by public and private actors in various security domains.
You will learn what “big data” is, and how big data relate to
algorithms, machine learning, and artificial intelligence. You will
learn the basics of the European human rights framework, consisting of
the Charter of Fundamental Rights of the European Union and the European
Convention on Human Rights, which originates in the Council of Europe.
Throughout the course, you will learn to identify which human rights are
at stake when public and private actors use big data to protect national
security and public security, and to analyze whether a certain big data
application violates human rights or is compliant with the human rights
framework.

The course has a strong focus on the rights to privacy and data
protection. You will learn the difference between those two rights, and
how the General Data Protection Regulation relates to the European human
rights framework. That said, privacy and data protection are not the
only rights threatened by big data, and the course accordingly looks
into other human rights, ranging from the right to human dignity, to
freedom of expression, and the right to a fair trial. For all these
different human rights, we will discuss relevant case law from the Court
of Justice of the European Union and the European Court of Human Rights
and assess how these cases, of which some were decided in a pre-digital
world, can be applied to big data.

For each lecture, you will read a selection of academic articles and
case law. The lectures are taught by several lecturers who all do
research in the field of big data and human rights. The course thus has
a strong connection to academic research and trains you to develop a
critical and analytic mind as well.

Today’s digital and information-dense society produces a massive amount
of data. Much of this data is generated by or related to human behavior
and can inform social scientists about societal dynamics. Examples
include social media data, parliamentary minutes, email collections or
collections of stories that have been made available in a digital
format. In this course, students learn about such digital trace data,
that are often voluminous, unstructured, and/or embedded in complex data
structures – we refer to such data as Big Data. Students learn about how
Big Data differs from data generated by traditional social science
research methods and the opportunities and challenges that Big Data pose
in present day society. They are introduced to R, a programming language
which they will use to gather and link data, and to make sense of these
data. Students learn about ways to analyze data derived from social
media such as forums or social network sites by using both computational
and interpretive approaches. The latter are key to Small Data, data with
meaning to individual citizens.

Research in Bioinformatics in its broadest definition concerns the analysis of informational processes within living systems with the help of computers. To do this succesfully, Bioinformatics actively uses and integrates contributions from areas such as Mathematics, Computer Science, Chemistry, Medicine and Biology. Bioinformatics has recently become one of the keywords in the life sciences as well as in Biotechnological and Pharmaceutical industries. Although in essence the field exists for over two decades and bioinformatics techniques developed over the years have come of age, the field has gained major prominence relatively recently, owing mostly to the world-wide human genome projects and subsequent structural and functional genomics initiatives.

Bioinformatics and Systems Biology is a joint degree Master’s programme from VU Amsterdam and the University of Amsterdam (UvA). The programme teaches you to combine molecular and cell biology, computer science and mathematical modelling to integrate vast amounts of biological data into a fundamental understanding of life at the molecular level.

During the programme, you will develop skills in scientific research and a high level of abstraction and quantitative thinking. You will learn to apply and combine rapid developments in different research fields and obtain an interdisciplinary mind-set with many transferrable skills.

Observations from biological high-throughput experiments allow us to
improve diagnosis and give a personalized treatment plan for patients.
However, integrating data from several sources and using this data for
predictions is non-trivial.

This is a theoretical and practical Bioinformatics course on
computational methods for Translational Medicine; we will focus on
Bioinformatics methods that are used to predict the clinical outcome for
patients and analysis methods to obtain deeper understanding of complex
diseases, by combining data from various high-throughput experiments
such as proteomics, microarrays and next-generation sequencing as well
as existing biological databases.

In the analysis of biochemical systems, many measurements are performed,
leading to complex multivariate data sets. The tendency is to measure
more and more of just a few samples. Multivariate data analysis methods
are often used to explore such sets.

This course covers a broad range of multivariate data analysis methods,
for example exploration, clustering, classification. The latter is
especially important in biomarker discovery. Design of experiments and
ANOVA for multivariate data is also discussed. Furthermore, the
interpretation of selected features in terms of function and networks is
discussed. The course starts with an introduction on the properties of
the different types of functional genomics data.

The main goal of this course is to teach students how to interpret the
results of the multivariate methods and how this relates to the
biological problem that is studied.

The purpose of this minor is to acquaint the student with the different fields of Neuroscience. The student will gain insight into the latest knowledge of how the brain works and also how this knowledge can be used to understand cognitive processes, social interactions between individuals, anti-social behavior as well as different brain diseases, such as depression, addictions, attention, or eating disorders. The nature-nurture debate will be discussed as well as recent updates in human genome research.

In addition, the minor provides an introduction into the fields of neuro-economics (decision making) as well as into recent scientific technological advances in brain-machine interfaces, deep brain stimulation, and robotics. The integration between disciplines, such as biology, psychology, sociology and genetics plays a central role in this minor. Students learn to think critically about how knowledge of the brain and the human genome can be applied to deal with societal issues.

Business Analytics (BA) is a multidisciplinary program aimed at solving quantitative business problems. The combination of mathematical and information technological methods plays an important role in this program, with the ultimate goal to improve and optimize business processes.

You are trained in identifying and solving problems in a very diverse field. Three disciplines, mathematics, computer science and economics contribute to such training. In addition to basic courses in these disciplines, a number of interdisciplinary courses and practice-oriented components are part of the study.

Characteristic for Business Analytics is its focus on the entire process of solving business problems. Starting with data you obtain insight in the underlying problems and formulate key business components with mathematical models. Such models will typically be analyzed and optimized using mathematics and computer science. You also learn how the solution can be implemented and what obstacles may play a role. In short, you keep working on all aspects involved in improving business processes.

In today’s data driven world, organizations need expert judgement and sharp analysis to ensure they make the right decisions. Business Analytics enables you to harness the power of data science, big data, statistics and machine learning to optimize results and achieve strategic objectives.

Your ability to combine insights from mathematics, computer science and economics with highly developed quantitative and communication skills will make you key to the success of any organization. The Master’s programme will deepen your knowledge in these areas and give you the opportunity to specialize in Computational Intelligence, Optimization of Business Processes, Financial Risk Management, or Research. The Master’s degree is concluded with a six-month individual internship at a company, which is often the first step to a thriving international career. In the Research specialization, a more research-oriented final project is an option, too.

If you are excited about subjects like artificial intelligence (AI) and computer science, our BSc Business Analytics could be for you. Nowadays, organisations deal with huge amounts of freely accessible and interconnected data. In this newly developed Bachelor’s you learn how to help organisations use this ‘big data’ to improve their performance. A profession that is in rapidly-growing demand on the job market.

In the minor Business Analytics & Data Science you deepen your knowledge on various aspects in the field of data science. The minor consists of a number of advanced courses in that field and complements the corresponding bachelor programme in Business Analytics in the direction of data collection and data-driven solution methods. The minor is concluded with the hands-on project Data Wrangling, in which you learn how to scrape data.

The Research Master Business Data Science is a multidisciplinary research program in which course instruction is provided by top scholars from the three participating Schools with a central focus on the performance of academic research within business disciplines, such as entrepreneurship and innovation, finance, human resources and organization, marketing, and supply chains analytics. The Research Masters prepares talented and motivated students to start their doctorate at one of the Schools in business and economics of the three partner universities or any other high quality PhD program in Business. A job market support program for PhD students will support BDS graduates to enter the international academic job market.

Dealing with the overabundance of data and the ability to transform data
into insights are critical success factors for organizations. This
course offers an introduction to the tools and concepts that allow you
to unleash the power of data and business intelligence and analytics
solutions in order to
create competitive advantage. The course primarily has a managerial
focus. The students will acquire hands-on experience with well-known
BI&A technologies, and learn how to use their features and capabilities
in practice. Our partners from the industry and the business consulting
sector will be closely involved in the course, sharing their insights
and experience during several interventions.

Keywords: business intelligence, analytics, data science, data
warehousing, data mining, big data, machine learning, data-driven
decision making

Dealing with the overabundance of data and the ability to transform data
into insights have become critical success factors for organizations.

This course offers the handles that are needed to unleash the potential
of data, and business intelligence and analytics solutions in order to
create competitive advantage. The course primarily has a managerial
focus. The students will acquire hands-on experience with trending BI&A
technologies to learn how to use their features and characteristics in
practice. Our partners from the industry and the business consulting
sector will be closely involved in the course, sharing their insights
and experience during several interventions.

Keywords in this area are ‘big data’, ‘data science’, ‘business
intelligence’, ‘data mining’ and ‘data-driven decision making and
innovations’

Business IT & Management is one of the learning paths of the HBO-ICT bachelor programme. In this programme you will gain the knowledge and experience needed to bring ICT and company strategy together. You are a networker and are able to easily communicate with stakeholders on different levels: managers, users and programmers. You combine these social skills with your knowledge of ICT, allowing your projects to run smoothly and in time. The resulting ICT applications will help companies to work more efficiently and more customer-friendly. Business IT & Management is one of the learning paths within the HBO-ICT bachelor programme. Other learning paths are Game Development (GD), Software Engineering (SE), System and Network Engineering (SNE) and Technical Informatics (TI).

This course is in Dutch.

During this course we study the different aspects of Monte Carlo
simulation and discrete-event simulation in a coherent way. Subjects
treated are: Modeling of business problems, statistical outcome
analysis, simulation optimization, software tooling, programming of
simulations in Java.

International business administration is a subject in which data are of
prime interest. Is there convincing evidence that your online marketing
campaign results in more sales than your standard newspaper adds? Or
that your increase in employees’ schooling budgets increase company
loyalty? As a business professional, you want to act on evidence, not
only on your gut feeling. Statistics is the key tool that helps you to
analyze and make sense of empirical evidence and to support informed,
data-driven decision making. These skills are highly valued in todays
labor market.

This course is part of your methodological toolkit and the
methodological learning line of your program. It builds on your aptitude
with symbols as built in Business Mathematics, and your critical
academic evaluation as trained in Academic Skills. The course also leads
up to a further deepening of tools in Business Research Methods, and
applications as dealt with in your courses on Business Processes,
Integrative Project, and the Bachelor Thesis.

Real functions of one variable. Topics that will be treated are:
1) Preliminaries, Real functions, Trigonometric functions
2) Limits, Continuity, Intermediate Value Theorem
3) Transcendental Functions, Inverse Functions
4) Differentiation, Chain Rule, Mean Value Theorem
5) Applications of Differentiation, Extreme Values, l’Hôpital’s Rule,
Taylor Polynomial
6) Integration, Fundamental Theorem of Calculus, Improper Integrals

Series, vectors in R^2 and R^3, real functions of several variables,
complex numbers, differential equations.
Topics that will be treated are:
1) sequences, series, power series, Taylor series
2) vectors, dot product, cross product, vector projection, distances in
R^3
3) partial derivatives, chain rule, gradients
4) extreme values, Lagrange multiplier method
5) double integrals, polar coordinates
6) complex numbers
7) first and second order differential equations

More information about procedures and registration periods can be found at http://student.uva.nl/sne/az/item/course-registration.html

Coding skills are increasingly in demand: they enable scholars to develop the appropriate applications for processing and analyzing data, either big or small.

This course teaches foundational coding skills using Python (a popular programming language), with the goal of:

1. helping students and researchers to understand when and how to automate a task or analyze data programmatically;
2. developing custom applications, rather than using ready-made ones, which can benefit the actual practice of humanities research as well as its outputs.

The course introduces foundational programming concepts (variables, data types, flow control, functions, input/output), and mentions useful extensions focused on data analysis (Pandas) and natural language processing (Natural Language ToolKit).

In this course you will learn about the theory of combinatorial optimization problems. Also, you will apply the theory to model and solve complex problems using the available software.
In particular, we consider performance measures for algorithms for combinatorial problems such as the running time and the quality of solutions.

In the study Communication and Multimedia Design (CMD) you will learn to design digital media with a view to the end user. This enables products, services and even entire organizations to improve.

This programmes is in Dutch.

Computational complexity theory deals with the fundamental question of how many resources (such as time, memory, communication, randomness, etc.) are needed to perform a computational task. A fundamental open problem in the area is the well-known P versus NP problem, one of the Clay Millennium problems. In this course we will treat the basics of complexity theory, including: NP-completeness, diagonalization, non-deterministic computation, alternation, Boolean circuits, randomized computation, approximation algorithms, subexponential-time complexity.

In the course Computational Intelligence, we will focus mainly on
computational aspects of Artificial Intelligence, namely, optimization
algorithms for solving learning problems. Specifically, we will consider
problems that cannot be solved using information about gradient due to
their combinatorial character or complexity of the objective function
(e.g., non-differentiability, blackbox objective function). These
problems pop up in computer science and AI, such as, identification of
biological systems, task scheduling on chips, robotics, finding optimal
architecture of neural networks. For this purpose, we will introduce
different classes of algorithms that can be used to tackle these
problems, namely, hill climbing and local search, and evolutionary
algorithms. Additionally, we explain sampling methods (Markov Chain
Monte Carlo) and population-based sampling methods, and indicate how
they are linked to evolutionary algorithms. In the second part of the
course, we will discuss neural networks as current state-of-the-art
modeling paradigm. We will present basic components of deep learning,
such as, different layers (e.g., linear layers, convolutional layers,
pooling layers, recurrent layers), non-linear activation functions
(e.g., sigmoid, ReLU), and how to use them for specific problems. At the
end of the course, we will touch upon alternative approaches to learning
using Reinforcement Learning and Bayesian Optimization. We will conclude
the course with a recently revived field of neuroevolution that aims for
utilizing evolutionary algorithms in training neural networks.

In this course, we discuss numerical and simulation-based estimation methods and their use in econometrics and data science. We start with a small recap of linear regression and discuss properties of the estimators. In the second part we discuss the assumptions made for these results and introduce a new simulation based hypothesis testing method called Monte Carlo testing. In the third part, we move to a more complex setting where less assumptions are made, and we discuss the foundations of bootstrap testing.

We illustrate how these methods are used in practice for a variety of econometric models including heteroskedasticity, autocorrelation and nonlinear models.

Computational Science is an area of science which spans many disciplines, but at its core it involves the development of models and simulations to understand natural systems. Our Master’s programme is designed with this in mind. We define three course types: compulsory core courses, restricted-choice courses and free-choice elective courses. A complete programme consists of 14 courses and a six-month graduation project.

Digitisation poses new questions and challenges for our societies, but it also brings new opportunities to intervene in societal issues such as inequality and climate change. Would you like to learn how to make the world a better place? Scientifically grounded in social sciences and humanities expertise, Computational Social Science students develop skills to design, create, apply and appropriate digital technology and data science to benefit society.

In the Master’s programme Computer Science you are taught by scientists who have a reputation for world-class research on computer systems, software, networking and online systems. As a joint degree programme of UvA and VU Amsterdam, you benefit from the expertise of both research groups at both institutions and their international contacts.

During your Bachelor of Computer Science, you will acquire a broad foundation of knowledge about computers and how they work. The lectures and assignments are challenging and are delivered by teachers who combine teaching with high-quality research. You will study mathematics and logic and learn to code in different languages. You will also immerse yourself in security and network technology.

Digital cameras have become ubiquitous in the form of consumer cameras, webcams, mobile phones, and professional cameras. These cameras yield enormous streams of data and provide the means for communication, observation, and interaction. In this course, image understanding is addressed with the focus on core vision tasks of scene understanding and object recognition.

A broad range of techniques are studied on how computers can understand the visual world of humans including image formation and filtering, features (color and shape invariants, interest point detectors, descriptors, SIFT, HoG), visual information representation (vector space, statistical models, bag-of-words), learning and classification (nearest neighbor, kernel density estimation, SVM), dimension reduction (PCA, LDA and SVD), object detection and classification, object tracking (mean-shift, Kalman), and user interaction (active learning).

This year, different advanced applications in human behavior understanding are studied such as face and emotion recognition, human body analysis and affective computing. Further, we concentrate on object recognition in the field of computer vision. We discuss the data, tasks, and results of Pascal VOC and ImageNet, the leading benchmarks. In addition, we discuss the many derived community initiatives in creating annotations, baselines, and software for repeatable experiments.

The field of computer vision is by impact one of the forefront fields of AI.

Computer vision has matured to a degree that many applications have become possible or nearly are possible.

In this course, computer vision is seen as an enterprise that uses statistical methods to disentangle image data using models constructed from geometry, physics, and machine learning.

The course aspires a thorough understanding of many of the current techniques, aiming at a broad basis, and extending to state of the art methods. Topics range from texture analysis, (photometric) stereo, and multiple view geometry, to sophisticated techniques of model based vision, material recognition, tracking, object detection, and scene classification.

To appreciate the nowadays possibilities, practical experience will be gained with some state-of-the-art techniques.

This course provides a comprehensive presentation of the foundations and
programming principles for multicore computing devices.

Specific learning objectives are:
* To provide insight into fundamental notions of multicore computing and
their relation to practice: locks, read-modify-write operations, mutual
exclusion, consensus, construction of atomic
multi-reader-multi-writerregisters, lost wakeups, ABA problem.
(Knowledge and understanding)
* To provide insight into algorithms and frameworks for multicore
computing and their application in multi-threaded programs: mutual
exclusion algorithms, spin locks, monitors, barriers,
AtomicStampedReference class in Java, thread pools in Java,
transactional memory. (Knowledge and understanding)
* Analyzing algorithms for multicore computing with regard to
functionality and performance: linearizability, starvation- and
wait-freeness, Amdahl’s law, compute efficiency gain of
parallelism.(Knowledge and understanding) (Applying knowledge and
understanding) (Making judgements)
* Mastering elementary data structures in the context of multicore
computing: lists, queues, stacks. (Applying knowledge and understanding)
(Making judgements)
* Programming in multi-threaded Java, and performing experiments with
such programs. (Making judgements) (Lifelong learning skills)

The importance of what is known as customer focus is widely recognized as key to success in the marketplace. Companies base their (marketing) strategy on their understanding of their customers. This course provides insight into how people behave as consumers and discusses the theoretical and managerial implications of such behaviour for firms. To do so, the focus is specifically put onto the innovation environment. This course takes a future-oriented perspective and uses current developments such as social media, “big data”, and a number of (marketing-relevant) disruptive innovations as application field to explain consumer behaviour.

We will cover topics such as capital structure, mergers and acquisitions, payout policy, equity offerings, as well as behavioral corporate finance. The course offers a rich blend of lectures, readings, exercises and cases. For all relevant topics specific references to practices and trends in in

The scope of corporate strategy goes beyond the individual product market, instead focusing on how firms can create value across different businesses. Thus, how can the company create value above and beyond the value created by the individual business units. In this course we will focus on three main decisions: Corporate governance (ownership vs. control and stakeholder strategy), Business portfolio choices (growth and vertical and horizontal integration), and organisation (The role of corporate headquarters, organisational structures).

More information about procedures and registration periods can be found at http://student.uva.nl/sne/az/item/course-registration.html

This course teaches how to apply computing technology to financial trading strategies. This is a broader set of skills than simply writing programs that execute trades. Students who complete this course will gain intuition into how financial trading strategies depend on relationships between different securities’ prices, or between prices and firm characteristics. Students will also learn that an algorithm’s fundamental purpose is to provide a set of detailed instructions for a computer to execute.

To achieve these goals, the course will demonstrate how to use the Python language to program various financial algorithms, such arbitraging price discrepancies between related securities or factor models that identify stocks with high returns. The first 1/3 of each lecture will explain the theoretical relationships behind each financial trading strategy. The other 2/3 of each lecture will demonstrate how to write programs that execute the strategies. Throughout the course, students will also learn the limitations of trading algorithms.

The programming section of each lecture will emphasize the importance of breaking down a trading strategy into a specific set of steps. It will then show how to solve each step using Python. Weekly in-class assignments will help students to gain practice with coding.

By the end of this course, students should possess a good understanding of the foundation of financial trading algorithms, and how to think like programmers. The goal is to provide enough knowledge of the fundamentals of finance and computer science that students can subsequently teach themselves how to program additional trading strategies.

One of the most important recent innovations in journalism is the increasing use of data. Often referred to as data journalism, we see a development of using computational techniques to make use of, for instance, massive sets of documents (e.g., leaks), or government data, provided via APIs or scraped from the web.

In short, the increased availability of digital data, fueled by the trend towards open governance or the use of online media, has opened new ways for journalists to discover and research interesting and relevant stories. While the use of data in journalism is not new, the amount of data and their digital nature require new skills from journalists. At the same time, audiences are demanding greater transparency from news organizations, and the news cycle is ever-more choked with content, both of which challenge journalists to use data in ways that are creative, compelling, transparent, and innovative.

This course combines discussion of these developments with practical skills training. Students will be introduced to the programming language Python, widely used for retrieving data from the web and analyzing both textual and numerical data. Additional topics include data visualization, finding stories in large amounts of data, and cleaning messy data.

The course is intended to introduce Data Mining Techniques to students
that are new to the field as well as to more experienced students. The
main aim is to gain a more practical perspective towards Data Mining
Techniques/Machine Learning. Lectures will cover more basic things for
those new to the field (general introduction into Data Mining, classical
algorithms such as decision trees, association rules, neural networks,
ensemble learning, etc.) and on top will discuss advanced topics
including deep learning, recommender systems, big data infrastructures,
and text mining. A number of successful applications in the area will
also be discussed. In addition to lectures, there will be an extensive
practical part, where students will experiment with various data mining
algorithms and data sets. The grade for the course will be based on
these practical assignments (i.e., there will be no final examination).

n this course you’ll build your own toolkit of useful programs with which you can read, transform and analyse data that you might find in various scientific areas. You’ll gain experience with professional development by implementing a final project for which you’ll define a problem based on a public dataset, design a visual data interface and implement a data pipeline to bring everything together.

Assistance is provided in the form of online Q&A with other students and our staff, as well as individual online or on-site tutoring. Optionally, you can join a study group to learn together. Final presentations are in public.

Leerdoelen

• De student is in staat om de gehele data science pijplijn succesvol uit te voeren. Dat wil zeggen: data verzamelen, een exploratieve data analyse, data transformatie, beide op basis van grondige statistische kennis, het kiezen van een passend predictief model, dat uitgebreid te evalueren, ook op ethische aspecten, en tenslotte de resultaten visueel aan belanghebbenden tonen.
• Kunnen voorbewerken van typische data science data: (zeer) grote hoeveelheden tekst met annotaties, gerepresenteerd in een database, spreadsheet of XML of als een verzameling tekst bestanden, voor wetenschappelijk onderzoek. Hieronder valt het transformeren, annoteren, categoriseren, classificeren en ordenen van data. Alles door middel van computers, en zo min mogelijk met de hand.
• Kunnen evalueren van de kwaliteit van door een computer programma bewerkte data.
• Kunnen uitvoeren van een exploratieve data analyse, visueel, en via statistische analyse, het normaliseren van ruwe data, het vinden van redundante informatie, en het evalueren van de kwaliteit van de ruwe  data.
• Ter voorbereiding op het afstudeerproject: het laatste deel van de empirische cirkel: “formuleer hypothese, operationaliseer die in termen van de beschikbare data, programmeer en verzamel en analyseer de resulaten en vertaal die terug naar de oorspronkelijke hypothese.”
• De student kan eigen werk en dat van anderen testen op correctheid, voorzien van een heldere geschreven beoordeling, en kan ingeleverd werk van zichzelf en anderen ordenen op kwaliteit.

This course is only available in Dutch.

In the one-year Data Science Master’s track, you will acquire knowledge of the theories and tools used in data science. We will teach you how to use these tools for working with data in different domains, such as Healthcare, Media and Communication, Smart City, Life Sciences and Digital Humanities. Graduates have an integrated view on the possibilities and development of data science in society.

Students will benefit from the strong collaboration with Amsterdam Data Science (ADS), bringing together leading researchers across the entire life cycle of data science, from expertise in machine learning and information retrieval to human computer interaction and large-scale data management.

This interactive programme offered by the Amsterdam Business School will be taught by experienced university lecturers and consultants, who will share their knowledge and expertise on the topic. There will be plenty of room for discussion and work-related questions.

Modules

Join 1 or all of the 5 hands-on Data Science Essentials modules, the next edition will start in 2021. You will learn the key concepts of working with data and analytics techniques and grow the skills to efficiently run data science projects, answering your organisation’s key questions.

Module 1: Python Programming Skills
Module 2: Data Cleansing and Visualisation
Module 3: Business and Soft Skills
Module 4: Applied Machine Learning
Module 5: Applied Optimisation

De module DSA1 heeft als doelstelling om de student de eerste beginselen van data science in auditing bij te brengen. Doel is het zelfstandig kunnen ontwikkelen van onderdelen van een datagedreven audit-aanpak, waarbij eenvoudige analysetechnieken in de praktijk kunnen worden gebracht. Na afloop moet de student in staat zijn zelf ideeën op dit gebied te ontwikkelen, de opgedane kennis in de praktijk toe te passen, om te gaan met onvolledige informatie over dit onderwerp en zijn of haar kennis met andere professionals te delen.

De stof wordt behandeld aan de hand van een casus, waarbij de data-gedreven accountantscontrole van een havenbedrijf centraal staat. Binnen deze casus lost de student eenvoudige vraagstukken op in de programmeertaal R in een speciaal hiervoor ontwikkelde Jupyter-omgeving.

De module sluit aan bij de Eindtermen met raakvlak stream ICT – variant Assurance uit de Eindtermen Accountantsopleidingen, voor zover deze niet in andere modules aan de orde komen.

This course is only available in Dutch

De module DSA2 bouwt voort op DSA1 en heeft als doelstelling om de student bij te brengen hoe uit de statistiek bekende exploratieve en confirmatieve analysetechnieken in een datagedreven audit-aanpak kunnen worden ingezet. Na afloop moet de student in staat zijn zelf ideeën op dit gebied te ontwikkelen, de opgedane kennis in de praktijk toe te passen, om te gaan met onvolledige informatie over dit onderwerp en zijn of haar kennis met andere professionals te delen.

De stof wordt behandeld aan de hand van een aantal cases, waarbij de data-gedreven accountantscontrole centraal staat. Binnen deze cases lost de student eenvoudige vraagstukken op in de programmeertalen R en Python in een speciaal hiervoor ontwikkelde Jupyter-omgeving.

De module sluit aan bij de Eindtermen met raakvlak stream ICT – variant Assurance uit de Eindtermen Accountantsopleidingen, voor zover deze niet in andere modules aan de orde komen.

This module is only available in Dutch

Na het afronden van deze module beschikt de student over kennis, inzicht en vaardigheden op de volgende gebieden:

• Data-driven auditaanpak
• Steekproeven en het testen van hypothesen
• Extract, Transform, Load
• Correlatie en regressie
• Classificatie
• Visualisatie
• Data-analyse in fraudeonderzoek
• Process mining
• Advanced analytics (big data, machine learning)

This course is only available in Dutch.

This course covers the basic theory of multivariate data analysis with a focus on the most relevant multivariate techniques, as well as their application to econometric data in computer lab sessions. The first two weeks provide a steep introduction to Python, NumPy and pandas, data scraping, cleaning and wrangling.  We then cover statistical methods in data science, where we will study in detail the theory and application of model evaluation, shrinkage methods, dimension reduction techniques such as principal component analysis, linear discriminant analysis, prediction by analogy and model averaging.

Big data refers to data that are more voluminous, but often also more unstructured and dynamic, than traditional data. This concerns, in particular, data-collection that draws on Internet-based data sources such as social media, large digital archives, and public comments to news and products. One of the big challenges is to derive information from these messy or unrefined data. We will focus on (a) acquiring and storing data (b) data wrangling: cleaning, transforming, merging and reshaping the data and (c) computer-aided exploratory analysis using robust methods. Students are expected to be interested in learning how to write own programs where off-the-shelf software is not available. Some basic understanding of programming languages is helpful, but not necessary to enter the course.

The course is concerned with data structures and the design and analysis of algorithms. We study several subjects from the book by Cormen et al: linear data structures such as stacks, queues, linked lists, tree-like data structures such as binary trees, binary search trees, balanced binary search trees, heaps, graph-like data structures, and hash tables. Further we study several sorting algorithms. some graph algorithms. string matching, and the programming paradigms divide-and-conquer, dynamic programming, and greedy algorithms. We consider the worst-case time complexity and in some cases the correctness of algorithms.

This project stretches of the whole semester and will provide students of both tracks with the opportunity to apply their gained knowledge to solve a complex problem in a real world project. The Project is founded on two pillars:

1. Experience and understand the creative process of developing an interaction environment as part of research into complex systems, with a particular focus on stakeholder research, user-research, data identification, context mapping, interaction design from agile development to a technologic prototype, and evaluation (validation).

2. Stimulating personal & professional leadership, via activities that improve team building and project management skills, and activities that contribute to one’s intellectual development, autonomy and employability. These activities are either organized by the students themselves, or are offered in the form of workshops. The aim, of this course is also to introduce students to a rigours application of academic skills, such as research question formulation, experiment design, and evaluation.

Within the 20 weeks of this projects the students work in groups of not more than 5 students and design, implement, and evaluate the interactive prototype  for a complex application. The application has to fulfil the requirements provided by the client, where the focus lies on the finding of a multi-disciplinary solution

• Understanding the client requirements, resulting in a requirement document;
• Development plan and schedule;
• Creative exploration of means of interaction available to address the client’s problem;
• Reflecting on the findings and decision making regarding the potential prototype;
• Conceptualising the prototype;
• Iterative implementation;
• Test;
• The findings will be presented in a final report and in the annual DSP conference end January, where selected groups present during the conference sessions and all groups present during the poster sessions.

The abundance of data creates opportunities for organisations to make better decisions and gain a competitive advantage. The challenge, for most business professionals, is to understand how to analyse this data but also interpret and incorporate it into organisational strategy. This course equips you with cutting-edge tools which enable you to turn data into business value.  In this course, you will learn the core principles of data analysis and gain hands-on skills practice. The course emphasises the practical application of various statistical and machine learning techniques to analyse different real-world business data using R software. You will also explore up-to-date machine learning techniques, including Decision Tree, Random Forest, and Regression Analysis to analyse and inform business decisions.

As this is the last course the students take before they start with the master thesis project the major aim is to work on reflective skills – thus this is a reflection course. Students will have to evaluate and reflect on impact on society of data and ubiquitous computing systems as distributed data driven services.

Students will investigate the interface between sensors, data, APIs, machine intelligence and societal interventions with practical application to people and real world problems. Research led activities on the course will be centred around applying theory in projects involving building and programming prototypes of remote sensing devices and physical data driven interventions.

Entrepreneurship is fundamental to generate value-added from innovation and it is an increasingly important subject for students and professionals, also in the context of data science. The growing complexity of the data science sector and its accelerating dynamics urge professionals to think and act in an entrepreneurial way. Due to the informatization of society, data about nearly everything emerge every day. By using data, searching for inferences and patterns, may entrepreneurs help to better support their new ventures.

This course is especially useful for ambitious students who want to demonstrate their ability to analyze data in order to pursue a career as an entrepreneur or who want to work at an entrepreneurial firm. During the last hundred years, entrepreneurial innovation is the main generator of jobs and welfare in Modern Society, the “true source” of national competitive advantage. Many Universities, Research Institutes and Research Departments of large Enterprises have adopted policies to stimulate the relationship between entrepreneurship and innovation, in the hope of facilitating economic growth.

During the course, students will learn basic knowledge about how to successfully launch a new venture and its underlying business idea. Working in teams, students will be requested to collect and analyze data in order to identify and validate an innovative business idea – representing a starting point for a potential start-up – to present in a “business pitch” and a “short final report” at the end of the course. Students must support their proposal with data analysis.

The course is based on three main pillars:  1) lectures; 2) tutorials; 3) coaching.

• Lectures: aimed to transfer academic knowledge in the field of innovation and entrepreneurship; based on a recommended book, selected academic papers and the students’ participation in speeches of industry experts in different entrepreneurial fields.
• Tutorials: aimed at the development of personal entrepreneurial soft skills, which may contribute to entrepreneurial success. Tutorials are based on interactive lessons, discussion of selected empirical cases and execution of specific assignments representing the steps that entrepreneurs follow in order to create a new business idea.
• Coaching (after teams’ presentations in tutorials): it is the time in which students have the opportunity to discuss their business idea with the instructors. During the coaching students will receive advice and instructions regarding the development of their business idea, their presentation and preparation of the final report (for example, how to identify a “minimum viable product”, implement a strategic analysis of industry competitors, writing a solid business plan).

The course is concerned with base principles and important aspects of relational databases. The course objective is to obtain a good knowledge and understanding of relational database systems. This includes the ability to develop good database models, and to query and update databases using SQL.

Data and databases play a central role in any information system from transaction processing to enterprise systems and, of course, data science applications. The purpose of this course is to offer a solid understanding of the core concepts in this area as well as an opportunity to apply these concepts hands-on in structured exercises as well as in a much less-structured ‘living case’ setting. These core concepts are based on the relational data model and data modeling as well as SQL -the de facto standard database language – combined with data visualisation and the design of metrics and dashboards. The course includes a significant practical part.

This one-week Deep Learning course covers theoretical and practical aspects, state-of-the-art deep learning architectures and application examples.
The lectures will introduce to students the fundamental building blocks of deep learning methods and the weaknesses and strengths of the different architectures. Students will learn how to tailor a model for a particular application. During tutorials students practice the theory using exercises and have the opportunity to ask for additional explanation for those parts of the material perceived as more difficult. Computer lab sessions aim at making the material come alive and train students in how the methods learnt in class can actually be applied to data. The lab sessions are meant to work on the assignments, such that the students automatically keep up with the material.

The summer course welcomes Master’s and PhD students, alumni, professionals in economics and related fields, who are interested in deep learning. The level is introductory, targeted at participants who would like to familiarize themselves with the topic, and acquire a good basis from which to approach deep learning potential applications.

Topics covered

• Introduction to Deep Learning (High-level definitions of fundamental concepts and first examples)
• Deep Learning components (gradient descent models, loss functions, avoiding over-fitting, introducing asymmetry)
• Feed forward neural networks
• Convolutional neural networks
• Embeddings (pre-trained embeddings, examples of pre-trained models, e.g., Word2Vec)
• Recurrent neural networks
• Long-short term memory units
• Advanced architectures (Densely connected networks, Adaptive structural learning)

Topics that will be covered in this course include:

• introduction to deep learning;
• neural networks and deep learning basics;
• convolutional neural networks;
• recurrent neural network;
• deep learning on graphs;
• deep generative models;
• deep reinforcement learning.

Deep learning approaches achieve state-of-the-art performance for many language technology tasks, such summarization, machine translation, and question answering. In this course we will cover several of those problems and the corresponding deep learning approaches.

The material is organized into the following  four parts:

Part 1: Words (Tasks: Lexical representations and word similarities)

Part 2: Sequence classification (Tasks: sentiment analysis and question classification)

Part 3: Next word-prediction and sequence labeling (Tasks: language modeling, named entity recognition, part-of-speech tagging)

Part 4: Sequence-to-sequence modeling (Tasks: Summarization, Neural Machine Translation, dialog/QA systems)

This course does not use a textbook and the relevant reading material, mostly consisting of research publication, will be distributed at the beginning of the course.

The minor Deep Programming elaborates on important principles, different paradigms, and modern developments in computer programming. Students have to choose five out of the six available courses (among these six, the Project Autonomous Driving has limited capacity. See details in the study guide course description).  Equational Programming is an advanced course on programming in the functional language Haskell. Compiler Construction provides in-depth knowledge on building compilers for translating source code from a high-level to a lower-level programming language. Secure Programming focuses on cryptography in software development. Concurrency & Multithreading teaches foundations and programming principles for multicore computing. Advanced Network Programming covers advanced concepts in network communication, beyond simple socket communication. This minor aims to turn students into highly skilled programmers and effectively prepare them for entering a Master’s program in Computer Science.

This course provides an introduction to the statistical methods of
social science research. Students develop skills to describe collected
quantitative data (descriptive statistics) and to test hypotheses about
variation between and within groups for statistical generalizability
(inferential statistics). Topics covered in this course include:
univariate and bivariate statistics (central tendency, dispersion,
association, and correlation), preparation and ‘reading’ of tables of
relationships between variables, sampling distributions, hypothesis
testing and calculation of confidence intervals, simple and multiple
regression, and reliability analysis.

What can organizations and brands learn from interactions with users on social media (e.g., Facebook, Instagram), websites or other digital environments? What insights can communication specialists gain from user behavior in the digital space? Can so-called big data hep them to gain knowledge about industry trends and future prospects for their organizations? Web behavior, social media interactions, and other digital traces left by people produce data that record, often in great detail, interactions between different stakeholders (including consumers, media outlets, members of political parties, etc.) and organizations. This overflow of data poses important opportunities to communication professionals. For example, how can we use such data to evaluate the outcomes of political, marketing, corporate or public health campaigns or help predict the success of communication efforts? At the same time, new challenges arise. Organizations are required to collect and process digital traces respecting the privacy of the user, as well as in an ethical and responsible way.

This course focuses on the process that communication professionals use to identify communication challenges that could be answered with digital analytics. In this course, we focus on the different steps of this process, including the gathering and understanding of the data, preparing the data for analysis, building models, and evaluating the effectiveness of the models in formulating actionable recommendations that address the challenges. We also discuss how to relate digital analytics to communication theories, as well as evaluate the privacy and ethical impact of the activities performed (e.g., ethical machine learning).

Throughout the course, students become familiar with some of the most common types of digital data (e.g., web and social data) and tools (including Google Analytics), learn how to analyze digital data, and evaluate their limitations.

The sources and objects studied in history, media, literature, art, and social sciences are increasingly becoming available in digital formats. The minor Digital Humanities and Social Analytics will train you in how to create and analyse different types of data collections, using tools for text mining, data analysis and visualization.

The courses include hands-on training, research internships in ongoing research projects, as well as theoretical reflection on the promises of ‘the digital’ for your own discipline. Practical computational training will sharpen your analytical skills and enhance your job opportunities in the future.

To organize this minor VU Amsterdam works closely together with the KNAW Humanities Cluster in Amsterdam, where students participate in cutting edge digital humanities projects.

In the Course “Digital Humanities in Practice”. you will be able to
apply your DH knowledge and skills to a real-world DH challenge . We
have teamed up with respected researchers and renowned institutes to
present you with a number of possible projects.

Digitale transformatie is de radicale organisatieverandering die het gevolg is van de opkomst van digitale innovaties zoals kunstmatige intelligentie, robotica, digitale platforms, innovatie-ecosystemen, blockchain, virtual reality en internet of things. We horen deze modewoorden overal, maar: wat betekenen ze nu echt voor jouw organisatie?

In deze opleiding voor professionals leer je navigeren door het digitale landschap om betere beslissingen te nemen. Zo herken je relevante kansen en bedreigingen en kun je middels digitale innovatie waarde creëren voor jouw organisatie. In de colleges bespreken we zowel de organisatorische als technologische uitdagingen van digitale transformatie, en leer je IT en business met elkaar te verbinden. De basis voor deze collegereeks zijn de nieuwste wetenschappelijke inzichten, tezamen met onze eigen onderzoekservaring vanuit het VU KIN Center for Digital Innovation over hoe digitale technologieën in de praktijk worden ontwikkeld en toegepast.

This course in only available in Dutch.

This course focuses on distributed computing systems and ecosystems. In
general, debugging and tuning existing systems and ecosystems, and
designing, implementing, and analyzing new distributed computing systems
remains vital and challenging for both industry and academia.

Starting with the mid-1990s, computing is undergoing a revolution, in
which collections of independent computers appear to users as a single,
albeit distributed, computing system. Motivated by the advent of the
Internet, by the increase in the computation capacity of consumer
computers, by the commoditization of server-grade machines, by energy
constraints, etc., the distributed computing paradigm has permeated all
fields using computers. Current distributed computing applications range
from social networks to banking, from peer-to-peer file-sharing to
high-performance computing used in research, from massively multiplayer
online games to business-critical workloads, etc. Important advances
have helped to fuse heterogeneous resources into truly global
distributed systems and ecosystems, for example in scientific computing,
where distributed computation is using Big Data and distributed sensors
to produce meaningful progress for the humankind. We will focus in this
course on a number of these modern examples of distributed computing
systems.

Although so many distributed systems and ecosystems already exist, the
list of conceptual and technical challenges they pose is long. Depending
on requirements, even trivial communication between nodes of the
distributed system can be challenging. The failure of a single node, or
sometimes even a performance hiccup, can bring an entire system down;
with it, other nodes or entire other systems may also crash,
experiencing correlated and catastrophic failures. Data consistency and
coordinating nodes remain important challenges made worse by the
large-scale of real-world deployments. Poor resource management and
naive scheduling can lead to orders-of-magnitude higher operational
costs and consumption of energy that we simply cannot spare. It is not
uncommon for a modern distributed system to quickly rise and then fall
in popularity, as exemplified by the 2016 example of Pokemon Go. We will
present in this course real-world situations where modern distributed
systems have behaved poorly.

Addressing these challenges requires unique approaches and concepts.
Separating concerns and breaking down problems into smaller cases often
lead to limited success, because many properties of distributed systems
can only be achieved end-to-end. Can anyone imagine a perfectly reliable
production pipeline, if even one of its key stages can suffer failures?
Building capability by adding resources is often offset by the
distributed nature of the system. Can anyone ignore the physical
limitations of communication around the globe? In this course, we will
focus on the unique approaches and principles of distributed systems,
from specific architectures and communication protocols, to specific
concepts in resource management and scheduling, data consistency,
fault-tolerance, and performance.

This course is concerned with reinforcement learning and its origin
dynamic programming. These are fields dealing with goal-directed
decision making over time, such as finding your way in an unknown area,
playing a game or pricing airline tickets.
We look at these areas from different angles:
– we deal with full-information “planning” problems, but also with
partial-information “learning” problems
– we consider different algorithms, some of which are guaranteed to find
the best solution, but also heuristics
– we consider high-dimensional problems (such as games) and methods to
solve them
– we look at small toy problems to understand algorithms and sharpen our
intuition, but also bigger problems for which we learn how to implement
algorithms (in python or R)
– we look at different types of applications, both from AI (search
problems, games) and OR

In this course you will be given an overview of the theory of discrete
and continuous dynamical systems (first period), as well as a foundation
in the most commonly applied numerical algorithms used to solve
algebraic and dynamic problems (second period) found in concrete
applications.

Dynamical Systems part:
1. Discrete-time dynamical systems: graphical methods to draw orbits,
calculating fixed points, stability analysis, period doubling.
2. Ordinary Differential Equations in 1 and 2 dimensions: graphical
methods, linearisation, phase plane analysis, classification of steady
states.
3. General theory of linear ODEs, solving initial value problems.

Numerical part
1. Introduction in Matlab programming
2. Finding roots of systems of nonlinear equations
3. Interpolation, Least Squares
4. Fast Fourier Transforms, analysing signals
5. Computing eigenvalues and eigenvectors, Pagerank
6. Numerical derivatives and integrals of functions.
7. Numerical methods for ODEs

The Master of Science programme in Econometrics is a multi-disciplinary Master’s programme providing a balanced and rigorous training in quantitative analysis of problems in economics and finance. The programme consists of advanced courses in mathematical economics as well as econometrics. Mathematical economics deals with mathematical modelling of market phenomena such as price adjustment processes. Econometrics deals with statistical modelling, estimation and testing whether economic models match observed patterns in real economic and financial time series. At the end of the Master’s programme, students are able to apply advanced mathematical and statistical methods, supported by modern software packages such as Eviews, R and Matlab, to explore and analyse problems in economics and finance.

The specialisation/track Econometrics emphasises statistical techniques for micro- and macro- econometrics analysis.

Good policy making requires understanding the impact of policies. How can we evaluate the effects of labor market policies? How can we measure the effectiveness of foreign aid programs? How can we identify which types of education are best for reaching learning goals? This course aims to answer such questions by offering an introduction to causal data analysis. We’ll cover a range of modern econometric techniques that are widely used in applied micro-economic research.

The first half contains basic econometrics, including linear regression models, hypothesis testing and coefficient interpretation. Considerable attention will be spent on omitted variable bias and selection bias, and how these concepts relate to the difference between correlation and causality. The second half moves through popular techniques that establish causality: Randomized Controlled Trials, Instrumental Variables, Regression Discontinuity and Panel Data models. Diverse applications in labor economics, health economics and development economics are discussed as illustrations. The focus of the course is applied rather than theoretical and students will make data assignments, in which they’ll learn how to implement these techniques in R.

Did the economic crises have an effect on elderly care? How can we increase profits sustainably? How will Brexit affect the European economy? Econometrists help to answer these questions by analysing real data through mathematical and statistical models. In the BSc Econometrics we train you to programme these models and interpret their outcome effectively.

In this course the multiple regression model is developed with applications, in particular, to cross sectional data. The treatment makes extensive use of matrix algebra and multivariate statistical theory. Discussed are: the classic linear regression model, standard assumptions, properties of the LS estimators, fit, consequences of omitted or redundant variables, partial regression, multicollinearity, linear restrictions, prediction, asymptotic properties and variable transformations, dummy variables, test for parameter stability, test for normality, heteroscedasticity, serial correlation, endogeneity of explanatory variables and instrumental variables. Applications and simulations are carried out with the software packages EViews and R.

The course explains a number of fundamental concepts that are important for the interpretation of quantitative results: including measurement errors, simultaneous equation bias, self-selection, censoring, truncation, etc. These ideas are fairly easy to explain with econometric models. Using these models, tests can be derived and solutions can be found for the problems involved.

Modeling itself is also an important skill and the course provides some initial techniques and extensions for correct modeling of economic variables. We will consider:

• non-linear methods including NLS and Maximum Likelihood;
• models with endogenous regressors (simultaneous equations, IV, GMM);
• models for discrete choices and other limited dependent variables (including Logit, Probit, Tobit);
• estimation and testing of the above models in practice using the computer;
• correct application of techniques and methods suitable for econometric research and the correct reporting thereof.

In today’s society, massive amounts of data are collected. But how is all that data used? How does KLM price its flight tickets based on supply and demand? How does Booking.com know why customers book certain hotels and not others? How does Spotify use algorithms to predict what its users want to listen to next? And what are the ethics behind collecting and storing all this data?

If you’re curious to find out, we’re curious to meet you.

The bachelor programme Econometrics and Operations Research combines various disciplines such as mathematics, statistics, informatics and economics. The programme is both broad and specific. Because of your knowledge of econometrics and practical skills, you will be able to provide solutions for issues in the field of economics. Your models for example will be able to determine the effects on employment of an ECB-initiated interest rate decrease, show the risk scenarios of a specific investment strategy, or help create the optimal planning for the deployment of trains. Econometrics and Operations Research at the VU is a small-scale education programme. You can always count on the personal help of senior students and teachers, and high-quality education. This course is in Dutch.

The Master’s programme in Econometrics and Operations Research is an academic programme focusing on the development and application of quantitative methods for analysing economic issues in a broad sense. The components of the Master’s programme correspond closely with the department’s research interests, which means that many of the latest scientific developments in areas like financial econometrics, logistics and game theory find their way directly into the teaching programme.

This is a course for the Duisenberg Honours Programme in
Quantitative Risk Management. It is accessible for outside students, if
they have sufficient background in probability, statistics, linear
algebra, econometrics and programming.

The course starts out with the theory behind common estimation methods
for linear, non-linear, or even non-parametric models. This knowledge is
applied to the study of factor, principal component and panel data
models. The second part of the course continues with a focus on time
series models with time varying parameters, both in a univariate as in a
multivariate setting.
Students are required to implement some of the methods in case
assignments using computer coding. We use Python as our standard
programming language, but students are free to choose some other
language if they prefer.

This is a course for the Duisenberg Honours Programme in Quantitative
Risk Management.
Following part 1 of this course, we proceed with studying econometric
methods for time-series data. We start with univariate and multivariate
linear time series models. We study their estimation and inference
procedures. Next, we consider non-linear time series models, In
particular, we consider models for volatility and financial risk. We use
this as a stepping stone towards general non-linear time series models
with time-varying parameters. We close with an introduction to state
space models and non-stationary time series models and their
applications within finance.

Students are required to implement the methods studies in case
assignments using computer coding. In this way, they develop hands-on
expertise with the methods and become familiar with both their potential
and their limitations.

Students are to choose the coding language they prefer from the set
Matlab, R, Python, Ox.

This master course places students in contact with recent research in
econometrics and data science. Instead of reading textbooks prepared for
the classroom, the students will read recent research articles,
published in scientific journals. To prepare the students for this task,
the course starts off with lectures that introduce the topics of the
research articles. Furthermore, exercises will be provided to deepen the
material discussed in the lectures and to further prepare for studying a
research article. A list will be provided from which students can choose
the research article they want to study. The list covers recent applied
as well as theoretical research. The students will also be trained in
writing and presenting research in econometrics and data science.
Students taking this course experience not only the difficulties and
challenges of research, but also the pleasure and excitement that comes
with it!

The Master of Science programme in Econometrics is a multi-disciplinary Master’s programme providing a balanced and rigorous training in quantitative analysis of problems in economics and finance. The programme consists of advanced courses in mathematical economics as well as econometrics. Mathematical economics deals with mathematical modelling of market phenomena such as price adjustment processes. Econometrics deals with statistical modelling, estimation and testing whether economic models match observed patterns in real economic and financial time series. At the end of the Master’s programme, students are able to apply advanced mathematical and statistical methods, supported by modern software packages – such as Eviews, R and Matlab – to explore and analyse problems in economics and finance.

Four specialisations are offered. Econometrics emphasises statistical techniques for micro- and macro- econometric analysis, whereas Financial econometrics focusses on mathematical and statistical techniques and their application to financial models and time series. Complexity and Economic Behaviour emphasises mathematical modelling of economic and financial markets. Data Science and Business Analytics deals with large and complex data from widely different sources for the use in economics and business. The specialisation depends upon the electives and Master’s courses chosen; a flexible mixture of these four specialisations is also possible.

In the course meetings we will discuss a range of approaches that are
often used to empirically analyse economics models, including instrument
variables, randomized controlled experiments, regression-discontinuity
designs, fixed effects models and differences-in-differences.

This course offers students the opportunity to study advanced empirical
research methods in finance. The objective is to increase the students’
ability to understand and to apply empirical methods in finance. The
course represents an integration of theory, methods and examples. We use
STATA as our standard software. The aim of the course is to enable
students to undertake their own quantitative research projects in
practice.
The course concentrates on the following methodologies: regression
models, panel data, endogeneity and instrumental variables, non-linear
models, logit / probit, credit risk, time series models, volatility
models (GARCH), forecasting.

This course covers key topics in contemporary empirical transport
research and policies. Key topics discussed (and applied in the
assignments) include:
– applied panel data analyses (e.g. estimate the effect of congestion or
market power on air travel delays)
– applied instrumental variables (e.g. estimate demand functions for
inland shipping and determine the welfare effects of low water in the
river Rhine)
– discrete choice analysis (e.g. to estimate value of time)
– discrete choice questionnaire and survey design
– using (estimated) models to analyse the effect of policies and natural
events (e.g. effect of a new rail link on mode choices and welfare)
– using (estimated) models to test hypotheses (e.g. do airline
‘internalise’ congestion they impose on themselves)
– dealing with data problems such a wrong entries, measurement mistakes
and outliers
– discontinuity designs, quasi-natural experiments’ and diff-in-diff
regressions
– (road/airport) congestion, global warming and other externalities and
the welfare and distributional effects of pricing them
– empirical studies of parking policy (e.g., cruising & taxation of
employer parking)
– company car tax policy and the effect on welfare
– the effect of public transport and road supply on road congestion
– competitive tendering in transport
– competition, market power and (de)regulation (of several types of
transport companies)
– lectures from national and international guest lecturers

After this course, the student should be able to:

– To understand the core concepts and models of entrepreneurship in both new ventures and large existing companies (intrapreneurship);

– To analyse and understand key challenges of innovation and launching new digital products and services including innovations organize to execute issues within larger organizations;

– To analyse how companies execute techniques from the start-up and venture world;

– To collaborate in a team and create and present a new offering that solves a real business need in a complex organisation, including a business model;

– Via a case study of GE industrial internet, learn how the largest industrial company in the world is turning themselves into becoming the Digital Industrial Company;

– About the ’transition gap’ – the phase between ’lean start-up’ and ’crossing the chasm’, a critical phase which prevents some start-ups from growing to their full potential.

In the practical work, we use the functional programming language
Haskell. We practice with the basics such as lists, recursion,
data-types, and a bit of monads.

The theoretical part is concerned with the foundations of functional
programming in the form of lambda calculus and equational reasoning. We
study untyped lambda calculus reduction theory and expressive power:
beta reduction, reduction strategies, confluence, encoding of
data-types, fixed-point combinators and recursive functions. In
addition, we study the lambda-calculus with simple types, its typing
system and a type inference algorithm, and possibly strong normalization
of simply-typed lambda-calculus. In the equational specifications part,
we study the syntax and semantics of equational systems, and we work
towards the results that all initial models are equal up to isomorphism
and that the term model is an initial model.

The process of data analysis consists of three phases: (i) data collection, (ii) querying the data and (iii) the consequences you draw from this analysis. Ethical and legal aspects play a role in all of these phases, and these will be discussed in this module.

Please note that this course cannot be followed separately.

The performance of a corporation depends on how well managers succeed increating shareholder value. We show you how to use tools that are offered by financial theory and help you just doing that: creating value. In this course we discuss three main issues in finance: capital budgeting, asset pricing and financial investments. The capital budgeting decision involves how firms select projects that create value. The theoretically optimal decision rule—th e net present value method—is discussed, also in relation to other selection criteria that are applied in practice. The asset pricing part concerns the way financial assets are priced by the market. The focus is on the pricing of shares issued by firms and bonds issued by firms and governments. Questions raised are: How are the term structure of interest rates and promised coupon payments related to bond prices? What is the influence of the expected stream of dividends and the level of market risk of firm’s projects on the price of shares? The financial investment decision is approached from a portfolio perspective and ends with a discussion of the Capital Asset Pricing Model (CAPM).

In this course, we aim to explain why accounting is seen as the ‘language of business’. Even though healthcare organisations often are not aiming to maximize their profit, like any organisation they need to explain their activities in financial terms. The field of accounting provides you with the tools and techniques to do so. Although it does not come natural to everyone, understanding the logic behind a unit cost, a balance sheet, or a profit number, will make your life as a manager easier.

The course deals with accounting techniques. These are not specific to healthcare situations. We will frequently use examples from a healthcare setting, but there is no such thing as a costing or budgeting technique for healthcare which uses a different logic or is based on other principles than those used for determining financial results of a company or the unit cost of a product. Having said that, upon completion participants should be able to understand and explain why it is difficult to evaluate the performance of hospitals from a financial perspective, or why it is difficult to improve financing mechanisms such as DBCs, DOTs or ZZPs.

This course covers the following topics:

• linear time series analysis;
• volatility models;
• value at risk;
• VAR models and cointegration;
• multivariate volatility and correlation models and high-frequency data and realised variance.

These topics are applied to empirical data using Python and R.

• Estimation and testing in the simple and multiple linear regression model;
• Violation of classical assumptions: autocorrelation, heteroskedasticity, specification errors, endogeneity, instrumental variables estimation;
• Prediction and forecasting, testing for structural change;
• Panel data models;
• Time series models: dynamic regression models, ARIMA models, Box-Jenkins analysis

Each lecture will be a mix of new material, discussing exercises, and live presentations in Python.

Unique two-day masterclass introduces participants to digital currencies, emerging mobile payment systems and blockchains, by the top expert of the world professor David Yermack from NYU Stern Business School.

This track aims at Computer Science students with a general interest in Computing and Concurrency and the application of formal methods for system design. Computing is a fundamental phenomenon in computer science and we provide courses addressing this field in a wide range: from distributed algorithms to protocol validation, and from term rewriting to logical verification. In order to enhance background knowledge and to support the further study of foundational questions some general courses in logic and mathematics are provided as well. Concurrency naturally occurs in the specification of distributed systems, and their analysis, verification and implementation require a systematic approach, aided by formal methods.

This courses focuses on the design and analysis of (multi) agent systems from a logical perspective. This course will be based on Michael Wooldridge’s book: An Introduction to Multi-Agent Systems (Wiley). The main topics include: goals and intentions, perception-action cycle, action planning, reasoning and search in agents, architectures for agent systems, collaborative agents, communication, and distributed problem solving. Focus will be on the logical approach to agent systems and on multi-agent settings. The main concepts will be applied to implementations. The course includes a practical part in which students implement agent systems and perform experiments.

Are you interested in Bioinformatics? Do you want to find out how
biology can make an exciting application domain? Or do you want to learn
what more you could do with your data, and with less effort? Enter here
to start!

Fundamentals of Bioinformatics (FoB) is the starting course of the
Bioinformatics master. It aims to give you a broad overview of important
topics relevant to the field, with a focus on current open problems. You
will be made aware of these open problems during practical sessions that
will cover active research in a major Bioinformatics problem (mutation
impact prediction). Based on your background, you will be assigned to
separate classes where you will be working to fill gaps in your
background knowledge in programming and/or biology.

Course Objectives [Dublin descriptors]:
* To make you aware of gaps in your own background knowledge [learning
skills]
* To address gaps in your background knowledge (programming, mathematics
and molecular biology) and skills [knowledge and understanding]
* To make you aware of the major issues when developing and benchmarking
new methodology in bioinformatics research [making judgements]
* To work together in a group of diverse backgrounds, to discuss and
report the results [communication]
* To apply and understand fundamental methods in bioinformatics, e.g.
sequence alignment, impact prediction, and to understand and use
fundamental concepts in bioinformatics, e.g. homology, evolutionary
conservation, and gene function [applying knowledge and understanding].

Data science is a dynamic and fast-growing interdisciplinary research field that, across science, industry, and government, is altering how people understand the world and make decisions. Not surprisingly, the demand for data science skills is on the rise. This course will cover key principles and tools of data science. In particular, the course will cover the process of acquiring and transforming data; the application of algorithms to learn from data (e.g., classification, regression, clustering); and the application of techniques to make decisions based on data, founded on introductory concepts of game theory and causal inference. The course will also cover the social and ethical implications of data science, with a particular emphasis on algorithmic fairness and explainability. The course will expose students to theory (i.e., machine learning and statistical methods underlying data science) and practice (i.e., use of data science libraries and analysis of real-world datasets).

During the course, students will work on a series of individual exercises and a group project that will bind together all elements of the data science process. Python will be used for all programming assignments and project. The course will introduce and make use of Jupyter notebooks, Numpy, Matplotlib and Pandas.

MAM 2 | Overview (act-e.nl)

Students will work in teams to perform experiments on software energy
consumption in a controlled environment. They will have to carry out all
the phases of empirical experimentation, from experiment design to
operation, data analysis and reporting. They will be provided with
examples of previous experiments, but they will have to choose by
themselves the experimental subjects and hypotheses to test. During the
lab sessions, students will be assisted for technical operation of the
lab equipment as regards measurement and data gathering. Students will
also receive the required training for data analysis and visualization
(i.e. graphs, dashboards) using specialized software.

Heb je altijd al interesse gehad voor (mobiele) apps, game development of opkomende technologieën zoals blockchain, big data, A.I. en virtual reality? Wil je nadenken over hoe robots ingezet kunnen worden in de zorg, of hoe bedrijven hun data beter kunnen beveiligen? Of wil jij met ICT-oplossingen bedrijven efficiënter laten werken? Dan is de Bachelor HBO-ICT misschien wel iets voor jou.

This course in only available in Dutch.

The course covers a number of key topics in the field of high performance computing and big data engineering. The course is organized as a lectures and workshops which help the students to develop both theoretical and practical skills. Following is the list of topics covers in the course:

• Introduction to parallel programming models and Big Data
• Cluster computing
• General-purpose graphics programing unit
• Machine Learning
• MPI/OpenMP
• Data management
• Local/Remote Visualisation for Data intensive application
• HPC Cloud

Researchers and engineers from industry and academia alike frequently experience their daily work to be impeded by physical limitations of their ICT equipment: processing and storage capacity, visualization facilities and their integration. They feel that up-scaling to high performance computing (HPC) facilities would be very beneficial to their work, but don’t know how to do this and lack the time to investigate their options.

The objective of this course is to introduce individuals with limited programming knowledge to various HPC facilities. At the end of the course, they will be able to use them avoiding common pitfalls, thus saving them money and time. The course is composed of a number of independent modules touching on various HPC and Big Data issues: Introduction to Unix, distributed systems, and Big Data; Using state-of-the-art Super Computers (with hands-on on the National Super Computer Cartesius and the Lisa cluster); HPC Cloud; GPU programming; Local and Remote Visualization Techniques; Data management; Data Intensive Computing with Hadoop: MapReduce and Pig; MPI/OpenMP approaches used in HPC and distributed computing.

This course gives an introduction to the relatively new field of ICT4D
and will be given jointly by the Department of Computer Science (CS) and
the Center for International Cooperation (CIS) with lecturers from both
backgrounds, covering different areas of expertise (social,
technological, organizational) in the field of ICT4D.

In the developed world computers are ubiquitous, and ICT has rapidly
grown into a critical asset for economic, technological, scientific and
societal progress.

The main objectives of this course are:

– To make the next generation of Computer Scientists aware of:
a) The importance of ICTs for the developing world and the unexpected
way
developing countries are leapfrogging into the information age.
b) The opportunities and challenges that exist for an information
scientist in
the area of ‘development4development’.
c) The influence of context in a typical ICT4D project.
d) The complexity of deploying an ICT project within a development
context, and how to tackle this and d) the benefits and pitfalls of
“Open Data for Development”.

– To equip the students with some initial project management,
technological and programming skills specific to an ICT deployment in a
developing country.

– Positioned at the heart of the VU’s vision of social relevance as one
of
the guiding principles, the core aim of the course is to raise the
awareness that we as Computer Scientists can make a significant
difference by sharing our expertise according to well established
principles of international development. Furthermore, this course will
give Computer Science students an opportunity to apply previously
acquired knowledge and skills in a specific application environment and
be able to transfer these skills to new application domains.

In the course we will give an overview over methodology, technology and
the social dimension of the usage of Information Technology in the
context of Development. We will introduce a general framework for
ICT4Development, we will teach you how to analyse a development problem
and introduce the analytical methods required for an in-depth
understanding of a potential development support project. Lecturers from
various backgrounds will provide some initial technological knowledge
required
for running an ICT project in a developing country. It will give an
overview over technology already applied, such as specific networks,
connection types, hardware as well as specific software environments,
but also introduce basic concepts in project management for ICT
projects. We will specifically focus on voice-based applications.

Sta jij er wel eens bij stil hoeveel je in aanraking komt met informatica? Of je nu naar Netflix kijkt, in- en uitcheckt bij het OV of je smartphone gebruikt: computersystemen zijn niet weg te denken uit het dagelijks leven en de ontwikkelingen gaan razendsnel. Ben jij zo iemand die graag uitvogelt hoe het allemaal precies werkt, zodat je het zelf nog beter kan maken? Bij de bacheloropleiding Informatica aan de UvA verdiep je je in de werking én ontwikkeling van computersystemen in alle facetten.

This course in only available in Dutch.

No organization can do without information systems. For some
organizations, such systems are even of strategic relevance, as they
offer a clear competitive advantage. Think, for example, of how Amazon
has become such a dominant retailer or how an organization as Uber has
conquered the taxi market.

This course explains the relevance and use of information systems in
modern organizations. We will briefly sketch how the role of information
systems has developed over the years to reach its current ubiquitous
level. Special attention is devoted to the rise of the internet and its
impact on traditional organizations, as well as the emergence of new
types of (cloud-based) organizations.

Reasoning from the organizational importance of information systems, we
will look into the way information systems are developed such that
organizations can achieve their objectives. We will pay considerable
attention to an important phase in information system development,
namely how we analyze and model business processes. For this purpose, we
will rely on the use of classical Petri nets.

This course will approach the topic of information management in breadth
and in depth. Breadth is achieved by giving an overview of all relevant
topics in the area of information management; depth is attained by
introducing students to a powerful, formal modeling technique that they
will learn to master in the context of organizational analysis.

This course covers the core aspects of information retrieval and search
engines, including indexing, Boolean retrieval, the different types of
queries, query execution, the vector space model, web crawling,
networks, link analysis, PageRank, classification, and clustering.

The course discusses state-of-the-art techniques that constitute the core of information retrieval (IR) systems, such as search engines, recommender systems, conversational agents, etc. The discussed techniques are divided into four pillars:

1. Evaluation: given a working information retrieval system, how do we determine its performance and how can we compare it to other systems?
2. Document representation and matching: given a collection of text documents, how do we represent these documents in a numerical form and how can we compare them to each other and to user queries?
3. Learning to rank: how do we adapt machine learning techniques to predict not only labels, but complete rankings of documents?
4. IR-user interaction: how do we model and utilize user interactions to evaluate and improve IR systems?

In the first part of the course, these four pillars are discussed for the most prominent type of IR systems, namely, search engines. In the second part of the course, the above techniques are discussed in the context of other IR systems, such as recommender systems, conversational agents, etc.

During the course, students are required to perform IR experiments. The goal of these experiments is to get acquainted with IR experimental methodology, to get a hands-on experience with state-of-the-art IR techniques and large datasets, and to be able to apply and adjust theoretical models to fit a task at hand. Besides running experiments, evaluating and analyzing the results is an important part of the practical side of this course.

Recommended prior knowledge

• Machine learning
  • A good understanding of machine learning algorithms is necessary (e.g., linear/logistic regression, naive bayes, svm, neural nets, trees).
  • An understanding and practical experience with estimation and optimization methods is also necessary (e.g., gradient descent, expectation-maximization).
• Information retrieval basics
  • A basic knowledge of how search engines are built and function is strongly recommended (e.g., crawling, indexing, text processing, vector space model and TF-IDF, precision/recall evaluation).
• Natural language processing
  • For most of the course, shallow NLP methods are in use. However, distributional semantics (e.g., word2vec) and topic models are important in the course.
• Software engineering skills
  • The course does not require to build large software or write immense amounts of code; however you should be ready to write code that implements algorithms discussed during lectures.
  • Python is the programming language of the course.

Building on IR1, the Information Retrieval course in year 1 at the MSc AI program, but also on courses such as Machine Learning 1, Deep Learning, and Natural Language Processing 1, this course will focus on the design, implementation, and analysis of state-of-the-art retrieval methods and beyond. An emphasis will be given to Neural Retrieval Methods and the applications of Question-Answering, Dialogue Systems, and Conversational Agents.

Retrieval and recommendation systems, are ubiquitous, to be found on many aspects of everyday life, ranging from product recommendation sites (e.g. Netflix, Amazon), to product search (e.g. bol.com), to web search (e.g. Google), legal search (e.g. Legal Intelligence), medical search (e.g. pubmed), digital assistants (e.g. Alexa, Siri, Cortana), and so on.

Over the years, IR ranking and recommendation functions have come to include features based on content-based analysis, based on the analysis structure (page, site, link), and based on user behavior. With the growth of computational power and the break-through in the field of machine learning (deep learning), novel architectures and neural networks have replaced the previous learning-to-rank models. With novel interfaces new means of communications between IR systems and users have been explored (e.g. conversational systems).

This course will be a team-based and project-based course, during which teams of students with the appropriate supervision, will be assigned to an IR project, they will conduct a systematic review on the topic, implement and apply state-of-the-art methods, and extend in novel ways. Project topics will focus on significant scientific areas in the field. In 2020/21 the topics will focus on question&answering, conversational search and dialogue systems.

This course aims to provide students with an understanding of how internal control has to be designed to realize reliable management information from a risk management perspective. Additionally the application of ICT technology within organizations is of great influence on the internal control of organizations. The formal organizational structure, segregation of duties and procedures can lose their importance if not completely redundant. The use of IT has consequences for the effectiveness and efficiency of the internal control and risk management systems of the organizations concerned. ICT requires continuous coordination between business objectives, governance, risks and compliance.

Information Sciences is the multidisciplinary area bridging Information and Communication Technology (ICT) and its practical use in society. Are you interested in how information is created and processed in companies and institutions? Are you more interested in the application of technology than technology for its own sake? Do you believe it’s important not to lose sight of the role people, organisations and cultures play in designing, modelling, communicating and sharing information? Are you fascinated by knowledge and innovation? If so, then the Master’s programme in Information Sciences at VU Amsterdam is an excellent choice for you.

Information studies is a broad and interdisciplinary field, primarily concerned with the analysis, collection, classification, manipulation, storage, retrieval and dissemination of information. It examines the interaction between people, organisations and any existing information systems, with the aim of creating, replacing, improving or understanding information systems. Information studies tackles systemic problems first rather than individual pieces of technology within that system: it focuses on understanding information problems from the perspective of the stakeholders involved, and then applying technologies as needed. Not only aspects of computer science are incorporated, but also aspects of research fields like cognitive science, commerce, communications, management, philosophy, public policy, and the social sciences. The Master’s programme in Information Studies at the UvA offers specialisations in: Data Science; Game Studies; Information Systems

Information Systems is the Master’s programme for you if:

– you are interested in the ways people interact with (new) technology and media, and how they are supported, hampered and influenced by them

– you want to analyse systems for the supply, storage and communication of information by means of various media (such as video, speech and text)

– you want to make connections between corporate/organisational management and the people responsible for developing technological solutions

– you want to translate user demands into innovative solutions

Information theory was developed by Claude E. Shannon in the 1950s to investigate the fundamental limits on signal-processing operations such as compressing data and on reliably storing and communicating data. These tasks have turned out to be fundamental for all of computer science.

In this course, we quickly review the basics of probability theory and introduce concepts such as (conditional) Shannon entropy, mutual information and entropy diagrams. Then, we prove Shannon’s theorems about data compression and channel coding. An interesting connection with graph theory is made in the setting of zero-error information theory. We also cover some aspects of information-theoretic security such as perfectly secure encryption, and draw some connections to machine learning and artificial intelligence.

The course will provide an introduction to some of the basic concepts of
Artificial Intelligence, such as search, adversarial search, knowledge
representation and machine learning.

Our international study trip is your chance to intensify your outlook on international business. During this exciting 1-week trip you take lectures at a top business school, visit local companies and networking events. During the study trip you will meet industry experts, executives, entrepreneurs and consumers in a classroom setting, but more importantly in real-life corporate settings.

In the last few years the destination of the International Study Trip was Silicon Valley, California, U.S., home of companies that have built their success on big data, like Google, eBay, Facebook and Apple. The programme offered insight into the nature of business in this leading hub for high-tech innovation and development. Students got the unique opportunity to personally experience the transformative power of big data and learn from real life cases.

Internet and the World Wide Web play a central role in our society, and have changed the way software systems are engineered and provisioned. Recent advances in virtualization techniques as well as the emergence of Software-as-a Service (SaaS) and cloud-based paradigms have enabled new ways of providing and exploiting computing and IT resources over the Internet. This track aims specifically at preparing students to work in such a complex, dynamic and distributed environment. It gives both in-depth understanding of the key components in developing distributed software- and service-based systems over the Internet, and provide the students with technical and critical thinking skills for the design and performance evaluation of such systems.

More information about procedures and registration periods can be found at http://student.uva.nl/sne/az/item/course-registration.html

Increasingly organizations and maybe even your future employer are
looking for experience as well as academic credentials. The School
recommends doing an internship, because it is an excellent way to apply
the knowledge and (academic) skills which you acquired during your
studies. Your most important learning goal as a student-intern is to
familiarize yourself with professional and market-related skills in a
real and new organizational environment. With the job market becoming
increasingly competitive, gaining relevant experience will give you a
good start into your professional career.

Companies offer a wide range of internships in various disciplines. What
is crucial in obtaining approval for your internship and eventually
obtaining your study credits, is that there is a clearly defined project
that allows you to fulfill the learning objectives. Also, the project
needs to allow for an individual assessment.

Finally, note that in order to obtain your internship credits, your
internship has to be pre-approved by the minor coordinator.

Orientation Econometrics:

When gasoline prices are increasing, more people will use trains to commute to their work. And when mortgage rates are decreasing, house prices will increase. These are apparently obvious relations between economic variables but are these consistent with reality, and if so, can we quantify these effects? These kinds of questions belong to the core of the profession of an econometrician. A research study is conducted by collecting useful data, which they then analyse using statistical models and methods to quantify the economic relations under investigation. The work of an econometrician consists of the empirical testing of economic theory, making forecasts and evaluating potential policy objectives. In order to be able to carry out these tasks, the researcher needs a thorough knowledge of statistical techniques. This course summarises these techniques and we will start with the application of it to a number of simple research problems. By using a couple of real life examples it will be explained what econometrics is all about.

Orientation Actuarial Science:

A life actuary analyses questions such as: how much should we pay for our pension benefits, and what should our retirement age be in the future? On the non-life side, actuaries analyse big datasets from insurance companies e.g., to construct risk-based premiums. Using risk factors in generalised linear regression models results in risk classification: which policyholders are good, and which are bad risks? It is critical for the actuary to be able to provide management with this type of information, based on advanced probabilistic and statistical techniques. Besides predicting the expected losses, actuaries are also interested in quantifying the uncertainty in the losses. Further, insurance companies and pension funds have huge assets under management. Simply investing these in the financial market can result in volatile asset returns. Derivatives can be used to hedge against these volatile movements, and the pricing of such derivatives will be illustrated.

Skills:

The Skills segment will be about conducting empirical research and writing a scientific research report. The topics which will be discussed are: the characteristics of academic texts, the use of a standard text scheme, the use of economic theory and relevant literature, the design of a research project, the presentation, and analysis of the results and the evaluation of the research project.

In this course, students get an understanding of what Business Analytics
is. You will work on a large case study that incorporates a variety of
aspects that you may find in a Business Analytics project. Moreover, the
course helps you to understand the contents and the objectives of the
Business Analytics curriculum. You will learn to see the connections
between the different scientific fields and the skills that you need.
You will develop data literacy, work with Excel for data analysis and
the construction of a mathematical model, and acquire an
analytic mindset that will help you in data-driven decision making. In
addition, 3 lectures are devoted to Academic Writing with corresponding
assignments.

B

Bij zien komt heel wat kijken! Visuele taken die de mens letterlijk zonder nadenken uitvoert (omdat ze in onze hardware zitten) kunnen we computationeel duiden en expliciet implementeren.

In deze cursus converteren we digitale beelden gebaseerd op pixels naar ‘visuele kenmerken’ (features) voor latere verdere verwerking in tracking, classificatie, of 3D-reconstructie. Voor die overgang gebruiken we wiskundige modellen van de lokale structuur en kleur van beelden, het menselijke visuele systeem, het afbeeldingsproces van de camera, en computationeel efficiënte algoritmen. Die modellen zijn de basis van computer vision, het zien met de computer! We slaan een brug tussen deze basis modellen en meer geavanceerde technieken zoals het categoriseren van plaatjes met convolutional-neural-nets.

Computer vision maakt gebruik van lineaire algebra en Taylor-reeksen van multi-dimensionale functies (de plaatjes), en heeft als doel deze te implementeren op de computer. In dit vak ga je dus het vak Lineaire algebra en het vak Continue wiskunde en Statistiek (met name de calculus) gebruiken en je moet kunnen Programmeren.

• Low level vision (interpolation, warping, local operators, convolutions),
• Local structure in images, Scale-Space, Feature Detection (SIFT),
• Pinhole camera, Camera calibration, Stereo Vision
• Motion, Optic flow, Tracking (als het blokschema dit toelaat)
• Convolutional Neural Networks voor computer visie

This course is avaliable only in Dutch

This course makes students familiar with standard microeconometric
methods. These methods are often used in economic research to test
predictions from economic theory. In addition, students also have basic
knowledge of algorithmic approaches, alongside more traditional
model-based approaches. During the course, attention is devoted to both
the theory underlying the different techniques and the practical
application. Theoretical knowledge is assessed in the final exam, while
the implementation of the different methods is assessed in an empirical
assignment. The software package Stata is used in the empirical
assignment. An important aspect of the course is that students learn how
to interpret estimation results.

Econometrics:
The goal of econometrics is to describe the relations between
observations as a useful model, from which predictions and inference can
be made. This introduction presents some first filters which can be used
for a
predictions, and continues with the principles of regression, applied to
real data.

Operations Research:
Core business in Operations Research is optimization. Several problems
from network optimisation are covered, among which the shortest path
problem, the minimal spanning tree problem, and the maximum flow
problem. For these problems, the mathematical structure is studied
leading to the design of algorithms for solving them. A glimpse will be
offered on complexity theory by analyzing the computation time of these
algorithms theoretically.

Mathematical Economics:
Within economic science frequent use is made of mathematical models.
Many of these models try to explain the choices of economic agents in
their economic environment. In this introduction, we set the first steps
in modelling mathematically the decision processes in economics, looking
both at behaviour surrounding individual choice in strategic decision
situations, and at cooperative decision making.

The course will comprise a basic introduction to MATLAB, designed to be suitable for students with little or no prior programming knowledge. The program will be based on introductory lectures and hand-on assignments, with the help and supervision of expert programmers. During the first half of the course the students will be guided to program their first tool to acquire actual neuroscientific data. In the second half students come up with a neuroscientific research question themselves. They will use their new knowledge to create a tool to acquire data to answer this question. Additionally, there will be introductory lectures on so called tool-boxes commonly used for data acquisition and analysis in neuroscience research.

In this course, students will learn how to program in the programming
language Java. Rather than focusing on Java-specific topics, the course
focuses mainly on the basic principles of programming, with emphasis on
good programming style and structure. These principles are applicable to
any object-oriented or imperative programming language, which makes it
easier for students to adapt to other programming languages during their
further research or study.

The following topics are covered: programming languages, algorithms,
primitive types, declaration statements, expressions, assignment
statements, object types, I/O (using the PrintStream and Scanner class),
methods, standard classes String and Math, control flow, writing
methods, writing classes, arrays, matrices, program design, recursion,
and using a graphical interface through a pre-programmed library.

During this course, students learn to write program in Python using
types (int, boolean, float, list and str), expressions, assignment
statements, if-statements, iterations (while- and for-statement), They
also learn standard functions, module math, as well as how to make
functions, perform I/O, make classes and use objects.

Introduction to Systems Biology is the starting course of the Bioinformatics and Systems Biology master (together with Fundamentals of Bioinformatics).

Goals:
– To make the student acquainted with the major approaches and methodology in systems biology (to be studied in more detail in the master).
– To develop a basic understanding of biological concepts that are relevant to current topics in systems biology.
– To gain hands-on experience in basic modelling as a means of solving systems biology problems.
– To repair gaps in background knowledge.

This course covers both theoretical and practical aspects of time series econometrics including the analysis of stationary and non-stationary stochastic processes in economics and finance. The students are introduced to autoregressive moving average (ARMA) models, autoregressive distributed lag (ADL) models, and error correction models (ECM).

Furthermore, the course provides both theoretical and practical insight into parameter estimation in time-series and the use of these models for forecasting, testing for Granger causality, and performing policy analysis using impulse response functions.

Finally, the students are introduced to the fundamental problem of spurious regression in time-series analysis. We find a solution to this problem by taking a journey into the theory and practice behind unit-root test, cointegration tests and error-correction representation theorems.

This course covers both theoretical and practical aspects of time series
econometrics including the analysis of stationary and non-stationary
stochastic processes in economics, business and finance.

The students are introduced to autoregressive moving average (ARMA)
models, autoregressive distributed lag (ADL) models, and error
correction models (ECM). Furthermore, the course provides both
theoretical and practical insight into parameter estimation in
time-series and the use of these models for forecasting, testing for
Granger causality, and performing policy analysis using impulse response
functions.

Finally, students become familiar with the fundamental problem of
spurious regression in time-series analysis. We find a solution to this
problem by taking a journey into the theory and practice behind
unit-root tests, cointegration tests and error-correction representation
theorems.

First, a review is given of least squares estimation and testing in the
simple linear cross-sectional regression model. We discuss the classical
assumptions, and the consequences arising when these assumptions are not
fulfilled. The linear model with multiple regressors is discussed using
matrix notation. Furthermore, we cover maximum likelihood estimation,
and models that are nonlinear in variables. Finally, an introduction to
panel data analysis is given. Throughout the course, the focus lies on
developing an intuition for
state-of-the-art econometric concepts. A balance is struck between
theoretical derivations and empirical applications. Extensive use is
made of the statistical software R, both for in-class illustration and
for hands-on exercises. An introduction to R is provided in the tutorial
of the first week.

In dit vak wordt ingegaan op de IT-infrastructuren die ten grondslag liggen aan onze moderne informatiesystemen. Aan de orde komen onder meer de eigenschappen van moderne datacenters, datacommunicatie en protocolfamilies als TCP/IP, dataopslagtechnieken, veel gebruikte platformen zoals Unix en Windows, cloud- en virtualisatietechnieken voor servers en werkplekken, database managementsystemen, middleware concepten als Service Oriented Architectures en messaging bussen en de integratie tussen de IT infrastructuur en applicaties. Beveiliging krijgt hierbij speciale aandacht.

This course is only available in Dutch.

The objective of the Knowledge and Data course is to make students
acquainted with methods and technologies used for expressing knowledge
and data, in particular on the Web. By the end of this course, students
will have built an intelligent web application that queries and reasons
over integrated knowledge from various sources obtained from the Web.
All of this will be based on formal logic theory.

Knowledge and understanding: at the end of the course, students will be
familiar with: Theory of Knowledge, Data and Information, Knowledge
Graphs, Semantic Web technology stack, Ontology Engineering,
Knowledge-driven Web Application Design

Application of Knowledge and Insights: students will be able to:
represent knowledge and data in various formalisms, implement basic
reasoning, develop advanced knowledge models and integrate acquired
knowledge in an intelligent semantic data driven web application.

Judgement: Students will be able to assess the value of available
datasets and ontologies for web applications, and to choose the
appropriate technology for a specific application.

Communication: Students are able to write a report about a developed
application.

Learning skills: The skill to acquire and apply knowledge and skills
about fundamental knowledge representation concepts as well as
state-of-the art technology, both individually as in a group context.

Knowledge Engineering is a discipline that involves integrating knowledge into a program for solving a complex problem, which requires human expertise. Typical tasks are classification, diagnosis, planning etc. In the course we use CommonKADS as the methodology for the process of modeling the organisation, the context and the knowledge intensive tasks. This methodology give clear guidelines and concrete templates for modeling the organisational aspects and the expertise model, which is the core model of knowledge based system. The notion of pattern-based knowledge modeling is a key issue in the knowledge modeling process. The goal of the final project is to perform the entire knowledge technology process for a knowledge intensive problem of your own choice, starting with context analysis, up to a (partial) implementation of the knowledge based system.

This course covers the general principles and methods that form the
foundation of information organization and knowledge-intensive
processes, as well as the contexts in which they can be applied and the
interaction with users. The lecture topics include knowledge modeling,
ontologies, logic, controlled natural language, Semantic Web and Linked
Data, as well as knowledge maintenance and evaluation, in addition to
guest lectures on specific applications and domains.

Knowledge and understanding: at the end of the course the students
should be acquainted with the broad principles of knowledge
representation, such as the separation of representation and reasoning,
the declarative nature of representations, the universal (domain
independent) nature of inference mechanisms.

Apply knowledge and understanding: students will have practical
experience with different representation
formalisms, and will be able to implement a reasoning tool for at least
one of these formalisms. This will allow them to better understand the
role of knowledge representation in the broader context of AI.

Making judgement: students be able to set up empirical experiments in
order to evaluate the pros and cons of Knowledge Representation
formalisms in specific application areas.

Communication skills: students will be able to write a scientific report
about an original research question in a small group of students.

Learning skills: students will be trained in acquiring knowledge about a
set of complex formal systems, learn how to come up with a research
question and scientific hypotheses, and perform the necessary
(empirical) research to prove or disprove those hypotheses.

In this course, you will learn the theory of knowledge representation languages that are used to express information on the Web, their application to real-world problems and data, and the research methods behind them.

The proliferation of data in all its forms has opened up the way to a new class of intelligent methods which are able to learn from this data. A large amount of this data is in human language form, either in written, or in spoken. This course will focus on technologies that allow machines to read and comprehend human language and generate human language themselves, so that human language can be transformed to information and communicated back to humans. A variety of applications will be presented aligned with tasks such as text clustering, classification, retrieval, question-answering, summarization, together with the underlying language technology which includes language representation, knowledge extraction and information retrieval.

This course confronts the students with some data management tasks,
where the challenge is that the mere size of this data causes naive
solutions, and/or solutions that work only on a single machine, to stop
being practical. Solving such tasks requires the computer scientist to
have insight in the main factors that underlie algorithm performance
(data access patterns, hardware latency/bandwidth), as well as possess
certain skills and experience in managing large-scale computing
infrastructure. Apart from the data being of large volume, another
problem invariably is that data comes in strange forms and formats, is
polluted, and needs to be questioned, transformed and cleaned.
The main part of the course is the second assignment: a large big data
analysis project where each student team tackles a different problem,
and while doing so gains experience in multiple aspects of large-scale
data engineering (critical thinking, data management technologies, cloud
infrastructure, visualization techniques, paper writing).

More information is found on http://event.cwi.nl/lsde – also check the
“showcase” section where you can see past project results
(visualizations and papers).

After completing this course, students should be able to:

• understand the (international) principles and values concerning privacy and (personal) data protection;
• understand the basics of the EU General Data Protection Regulation (GDPR) and related EU-legislation;
• be aware of the possible impact, threats and risks of big data for privacy;
• understand that not all data may be necessary to use and that technologies (such as anonymisation and pseudonymisation) of data can help avoid or minimise the impact, risks and threats for the data subjects and the parties processing the data;
• be aware of the available methods and standards to design privacy-friendly systems and services (Privacy by Design);
• understand the (international) principles and values concerning use of AI;
• understand the OECD and EC guidelines on using Artificial Intelligence;
• understand related concepts of transparancy, explainability and fariness of AI;
• understand how these concepts can be ensured in the context of organisational use of data and AI, applying a legal and ethical perspective to data and AI in real-life cases;
• form a view/opinion of an organisation, a customer and a government/regulator on the use of (big) data and AI.

Amongst emerging technologies robotics and artificial intelligence are
prominent both in terms of existing as well as expected use in society.
These technologies are special, because they come close to how we humans
function. At this moment both robots and artificial intelligence are
primarily used for specific tasks (playing games, surgery, self-driving
cars), but developments are moving fast. What exactly the future brings
is difficult to tell, but no one denies the potential and risks related
to robotics and artificial intelligence. Not surprisingly, in the legal
and policy arena an active discussion is going on related to legal and
ethical issues. These are the issues addressed in this course. The
legal angle includes both existing law and the need for new law. If new
law is needed, discussion will also be on how this new law should be
drafted. For instance, presently the European Parliament is analyzing if
maybe some time in the future we may need some sort of legal personality
for robots, and Harari is even fantasying about legal personhood for
algorithms. Ethics can apply to both the development and use of robots
and artificial intelligence. In this course ethics is primarily used to
either constrain the application of existing law or to guide the
drafting of new law.

Applications that are covered in this course include softbots, the
internet of (robot) things, ambient technology, autonomous intelligent
vehicles, and social robots (care and sex).

People are central to the success of organizations, but they are also the most difficult asset to manage. Therefore, managers require a strong understanding of how to lead and manage individuals and groups in order to effectively solve organizational problems and maximize long term performance. As a manager, you need to develop the skills and talents of employees, and motivate them to achieve strategic goals. You also need to effectively combine employees in teams, enhance collaboration and reduce conflicts. In addition, you help to build, strengthen, or change the organizational culture. You need to make things happen, and often under challenging and changing conditions or timeframes. In order to do this successfully, managers need to be able to diagnose and analyze problems, make effective evidence-based decisions, influence and motivate others, and manage their teams. This course aims to prepare you to make more effective managerial decisions and increase your impact as a manager.

This course is an introduction to the key elements of leading and managing people and focuses on frameworks, theories and tools that help to understand how to lead and manage people effectively. At the end of this course, students are expected to be able to:
Understand and explain important theories, frameworks, and research evidence related to leading and managing people effectively
Apply these theories, frameworks, and research evidence to people-related business problems
Diagnose and analyze problems, and make effective evidence-based decisions as a manager

This course concerns robots that can adjust and improve their behaviour
over time.

The course has a strong hands-on flavour. After two introductory
lectures students have to develop and implement the learning method of
their choice in simulation. In particular, adequate robot controllers
have to be learned autonomously for two tasks, maze navigation and food
collection. After testing and tuning the methods in simulation, the best
learned robot controller must be ported to a real Thymio and the real
world performance compared with that observed in simulation.

Vast amounts of data have been collected through genomics initiatives. They provide a golden opportunity to research the secrets of life, to understand more of its complexities, to improve quality of life and to conquer major diseases. Converting this huge volume of data into real understanding is the basic challenge of Bioinformatics research.

After successfully completing this course, the student
– has a working knowledge of the concepts of matrix algebra and finite-dimensional linear algebra, such as echelon form, LU-decomposition, linear independence and determinants;
– is familiar with the general theory of finite-dimensional vector spaces, in particular with the concepts of basis and dimension;
– is familiar with the concepts of eigenvalues and eigenvectors, diagonalization and singular value decomposition and can apply these concepts in basic applications in
discrete time dynamical systems;
– has working knowledge of the concepts of inner product spaces and matrices acting in inner product spaces, including orthogonal projections and diagonalization of symmetric matrices.

Logic is an interdisciplinary and international two-year Master’s programme at the University of Amsterdam (UvA) that focuses on the central role of logic as a mediator between the sciences and the humanities. It is an incredibly complex and fascinating programme, intended for students who want to relate traditional fundamental research in the formal sciences to a wide variety of applications, ranging from Information Sciences to Linguistics and Philosophy. It is a programme for highly motivated students from all over the world, who are able to work in a inspiring and demanding environment, both individually and in groups.

The focus of the lecture is on propositional logic and first-order
predicate logic. We work with natural deduction as a proof system. The
relation between semantic and syntactic methods is important; the
central keywords are correctness, consistency and completeness.
Moreover, we pay attention to expressive power, for example when
formulating queries. A fundamental tool, for this purpose, is the
compactness theorem.
Algorithmically there is the contrast between the decidability of
propositional logic and the undecidability of predicate logic (for
example, seen by a coding of the Post Correspondence Problem).
As a variation of the mentioned logics, we consider modal logic with
Kripke models as semantics.

The course Logistics Analysis is an exciting course that will challenge
you in various ways. By taking Logistics as a point of departure, we
bring together several perspectives and analyze business problems faced
by logistics companies. Taking a logistics perspective will stimulate
you to think about organizations in a different way, bringing together
knowledge from different fields and realizing that this creates
challenges and conflicts that managers need to deal with. You will learn
to systematically describe logistical systems, and identify problems
that emerge in these systems. Moreover, this course offers you a number
of tools that allow you to analyze logistical systems, optimize them,
(re)design them and assess the consequences of suggested improvements.
Important topics such as production management, inventory management,
and maintenance management are addressed, which are essential, hands-on
tools any logistics professional should be able to work with.

Course Objectives:
The overall objective of the course Logistics Analysis is to furnish you
with knowledge and skills to analyze logistics problems in a variety of
industries including manufacturing, transportation and services. More
specifically, we expect that upon completion of the course you are able
to:
– Understand and explain (the importance of) logistical systems and
their complexity;
– Analyze, model, and optimize real-life logistical systems by applying
quantitative tools;
– Interpret and discuss the use of these tools in specific logistics
situations as well as their general strengths and weaknesses;
– Write a clear management report containing useful recommendations for
the management of a logistical company;
– Clearly report your work: analysis, modelling, solution approach,
computational results, and managerial insights.

In the first year you will follow a general introduction in logistics. You will learn to produce what a client demands, how to distribute specific goods to a desired location, and you will learn what to buy to produce what a client demands. You will learn about the different disciplines involved in these processes, such as marketing, distribution logistics, production logistics and procurement logistics. Additionally, you will follow courses on mathematics, ICT, English, business administration and serious gaming. About a third of the programme will be given in a project-oriented format.

In the first year you will quickly get to know the logistics sector by doing project-based practical assignments and visiting companies in the field. In the second year you will further deepen your knowledge of logistics, such as setting up warehouses, managing the distribution of goods, tracking and tracing, and sustainable logistics. You will also go abroad on a study tour. In the third year you will follow a minor. You can – for example – choose to participate in the Logistics research programme, which focuses on Airport, Seaport or City Logistics. In the fourth year you will follow the final courses that will further sharpen your knowledge, and you will end the programme by doing a graduation assignment within a company. For ambitious students looking for an extra challenge we offer an additional education programme. This course is in Dutch.

The proliferation of data in all its forms being it symbolic, numeric, textual or visual has opened up the way to a new class of intelligent methods which are able to learn from this data. Applications of machine learning are broad and diverse and range from prediction of health parameters, understanding the content of social media, to recommendation of products. In this course we teach you the theoretical foundations of machine learning and how to apply these methods in practical analytic tasks.

Machine learning is the discipline that studies how to build computer
systems that learn from experience. It is a subfield of artificial
intelligence that intersects with statistics, cognitive science,
information theory, and probability theory.

Recently, machine learning has become increasingly important for the
design of search engines, robots, and sensor systems, and for the
processing of large scientific data sets. Other applications include
handwriting or speech recognition, image classification, medical
diagnosis, stock market analysis and bioinformatics.

The course covers a wide variety of machine learning techniques, but
puts particular emphasis on gradient descent optimization,
backpropagation, neural networks and deep learning. Some discussion on
the broader social impact of machine learning technology is included.

Upon successful completion of this course, the students will be able to:

• discriminate between different machine learning and pattern recognition methods, explain their main characteristics and choose an appropriate one for a given problem;
• apply the methods on different types of data;
• evaluate the performance of methods using different metrics.

• The student can explain, motivate and distinguish the main areas of machine learning in general and on examples
• The student can explain the major statistical learning frameworks/principles together with their advantages and shortcomings
• The student knows the major linear and non-linear statistical models together with their advantages and disadvantages, can explain them and the made model assumptions, can reason about and inside them, and can manipulate them
• The student can set up learning objectives with the taught models, can train and evaluate them and can assess the quality of fit
• The student can implement all the above in Python and apply the learned principles and models to real world problems and data sets

The main big two themes of this course are probabilistic graphical models and inference in various forms.
This course continues where Machine Learning 1 stopped. We will treat chapters 2, 8, 9, 10, 11, 13 of the book Pattern Recognition and Machine Learning by C. Bishop, plus some additional material made clear on Canvas. Topics include:

• Exponential families
• Conditional independence
• Information theory
• Independent components analysis
• Graphical models
• Latent variable models
• Learning, exact and approximate inference
• Variational Inference
• Sampling methods, MCMC, etc.
• Sequential data models
• Probability divergencies
• Causality

This course is a mix of machine learning theory in regular lectures and application of this knowledge on large datasets in practical sessions.

Topics include:

• Bayesian inference;
• Linear classification models;
• Neural networks and deep learning;
• Bagging, boosting and random forest;
• Machine learning methods for causal inference (e.g., causal forest, double-lasso, debaised-ML)

Machine learning originates from computer science and statistics with
the goal of exploring, studying, and developing learning systems,
methods, and algorithms that can improve their performance with learning
from data. This course is designed to provide students an introduction
to the main principles, algorithms, and applications of machine
learning. It includes topics related to supervised learning algorithms
for classification problems (logistic regression, support vector
machine), for regression problems (ridge regression, LASSO), but also
unsupervised learning algorithms (k-means, clustering, linear and
nonlinear dimensionality reduction). We adopt principles from
probability (Bayes rule, conditioning, expectations, independence),
linear algebra (vector and matrix operations, eigenvectors, SVD), and
calculus (gradients, Jacobians) to derive machine learning methods. We
further discuss machine learning principles such as model selection,
over-fitting, and under-fitting, and techniques such as cross-validation
and regularization. In case work we implement appropriate supervised and
unsupervised learning algorithms on real and synthetic data sets and
interpret the results.