Introduction to Machine Learning (UU)

The goal of this course is to give an introduction to machine learning. We will introduce several successful methods for regression, classification, clustering, and data dimension reduction and develop the mathematical theory behind these methods, with a focus on continuous optimization theory. In the exercise sessions, the students will work on proof-style exercises, as well as implementation exercises in Python.

The course builds on knowledge acquired in the first year, in particular Calculus and Linear Algebra 1 and 2, Analysis, Probability (or its predecessor Introduction to Probability and Statistics) and Computer Programming for Mathematics.

This course is one of the modelling courses in the bachelor programme with major ‘mathematics’ or ‘mathematics and applications’. Please find more information on the students website.

Program/Minor

Bachelor Computer Science (INCA); Mathematics (WISK); Mathematics and Applications (WSKT); Programma: Major Mathematics and Applications; Minor Mathematics (WISMINOR); Minor Mathematics (WISMIN);

Learning Outcomes

Basic machine learning terminology: different types of tasks, loss functions, training-validation-test split, cross-validation, generalization error, under- and overfitting, interpolation; Linear regression and regularization (ridge regression and the Lasso); Logistic regression; Support vector machines; Kernel methods; Decision trees and random forests; Introduction to neural networks; Data dimension reduction methods (e.g., principal component analysis and autoencoders); Clustering methods (e.g., k-means and spectral clustering); Gradient descent and its variants, e.g., stochastic gradient descent and accelerated gradient descent; Convergence analysis of descent methods; Theory for constrained convex optimization problems, in particular the Karush-Kuhn-Tucker theorem and Lagrangian duality.

Prerequisites

Calculus and Linear Algebra 1 and 2, Analysis, Probability (or its predecessor Introduction to Probability and Statistics, Computer Programming for Mathematics

Details

Institution: UU

Period: Period 2

Level: Advanced

EC: 7.5 ec

Introduction to Machine Learning 1

The core of this course revolves around programming machine learning algorithms for yourself, as a way to truly understand what exactly they are learning. Each of the different modules will focus on programming a different algorithm, understanding the math required for that algorithm, and discussing a philosophical question or a societal impact related to applying this algorithm in practice.

Specifically, we’ll cover the following algorithms: k-Nearest Neighbours; Naive Bayes; Gradient Descent; (Multivariate) Linear Regression; Polynomial Regression; k-Means

Note: This explicitly does not include Neural Networks, as that is too large a topic to also include here. However, the foundational concepts covered are also all applied within Neural Networks, and so this does provide the necessary basis to study Neural Networks in a follow-up course (IML2).

Program/Minor

Minor Machine Learning

Learning Outcomes

Know the main types of machine learning algorithms, and when to apply each; Know how to train a simple machine learning model using your own data set; Know how to detect underfitting or overfitting, and select the best model complexity for a task; Understand the foundational mathematics that allow all ML algorithms to learn from large amounts of data; Understand some of the important societal risks when applying ML in practice; Understand what it means for an algorithm to “learn” something, and the capabilities and limitations this implies

Prerequisites

A large part of this course consists of programming machine learning algorithms for yourself, so it is absolutely necessary you can write basic Python code. This includes writing your own functions, using list, dictionaries and tuples, and debugging a solution that doesn’t work. The course will also cover the mathematical foundations of these machine learning algorithms. This will require a very basic understanding of calculus, linear algebra, probability theory and statistics. Optional self-study modules for these topics are included in the course. If you need to brush up all of the mathematics required for the course, expect to spend an additional 4 hours each week.

Details

Institution: UvA

Period: Period 1, Period 4

Level: Intermediate

EC: 6 ec

Introduction to Machine Learning 2

In this course we continue with discussing machine learning models and algorithms. While the focus in Introduction to Machine Learning 1 was on programming basic models yourself, in this course we will make more use of libraries for the elementary parts and the focus will mainly be on how to combine these parts into more complex models. You will apply these more complex models in your own code to real data sets, where we will also cover common operations for such data.

Specifically, we’ll cover the following topics: Logistic Regression; Feedforward Neural Networks and Backpropagation; ReLU, Softmax, Dropout and Batchnorm layers; Decision Trees and Random Forests; Convolutional Neural Networks

Program/Minor

Minor Machine Learning

Learning Outcomes

You are able to use machine learning libraries to solve AI problems based on real data; You are familiar with several commonly applied machine learning models, such as random forests and (deep) neural networks; You can make and motivate a choice for a specific model based on a problem description; You can evaluate a learned model based on training and validation results, and based on this analysis make a suggestion to improve the model; You can determine what the hyper-parameters are that influence the complexity of a specific model and determine a strategy to optimize these parameters; For a specific model, you can indicate the exact function or value that is optimized by the model, how that optimum is found and what the possible limitations are.

Prerequisites

You should be familiar with all the topics covered in Introduction to Machine Learning 1. See the course catalog of IML1 for a complete list of topics. If you did not follow IML1, you'll need to have followed a different introductory ML course to meet the entry requirement. If you're unsure if a course would meet the entry requirement, you can contact the course coordinator beforehand. A large part of this course consists of programming machine learning algorithms for yourself, so it is absolutely necessary you can write Python code. This includes writing your own functions, using list, dictionaries and tuples, and debugging a solution that doesn’t work. In addition, you should be comfortable defining your own classes (OO programming), and using the following libraries: numpy, pandas, and matplotlib. Specifically, you should be able to use matrix multiplication and other linear algebra tools to vectorize code.

Details

Institution: UvA

Period: Period 2, Period 5

Level: Intermediate

EC: 6 ec

Introduction to Responsible AI Engineering

Introduction to AI and Responsible AI Engineering (hereafter Responsible AI for short) introduces students to the history of AI development and how it got entrenched in society, as well as to the range of ethical issues that tend to arise during the development, design and application of AI in society.

Program/Minor

Electrical Engineering, Mathematics and Computer Science

Learning Outcomes

Explain the main philosophical differences between strong and weak AI; Discuss different philosophical perspectives on technologies; Explain a range of views on the relationship between AI developments and society; Identify ethical features of specific AI-based technologies; Holistically approach environmental aspects of AI development; Apply philosophical concepts and approaches to the current debate surrounding responsible AI; Formulate a position on the ethical issues at stake in current AI applications regarding responsible design ideas; Write an academic paper with a concise research question, a clear structure, a justified choice of a theoretical framework, reference to academic literature and a consistent line of argumentation; Reflect on the structure and academic quality of the work of own peers in a deep and specific manner.

Prerequisites

The course does not require any prior knowledge in philosophy or ethics of technology.

Details

Institution: TU Delft

Period: Period 2

Level: Advanced

EC: 5 ec

Keystone Project III: Big Data

Welke genen worden geactiveerd bij stress? Waarom bevorderen sommige bacterieën immuniteit? Hoe ontstaat kanker? Hoe wordt een microbe een ziekteverwekker? Om dit soort vragen te beantwoorden worden er vaak grote datasets –“Big Data” gecreëerd. In deze gevallen is analyse van de resultaten met eenvoudige middelen (Excel etc.) niet meer werkbaar.

Om hiermee om te gaan gaat de student leren hoe Big Data gebruikt kunnen worden in het oplossen van biologische vraagstukken (experimental design), welke statistische analyses er gedaan kunnen worden en hoe de resultaten hiervan gevisualiseerd kunnen worden. Studenten leren omgaan met websites en eenvoudige R-scripts die analyse, modellering en visualisatie mogelijk maken. Deze skills worden aan de hand van biologische voorbeelden in de praktijk gebracht (computerpractica en zelfstudie) en door de student in een practicum verslag weergegeven.

Studenten zullen deelnemen aan projecten die verschillende typen biologische data gebruiken. Transcriptomics (RNA-seq) data bijvoorbeeld, dit kan inzicht geven in welke processen die als reactie op een stimulus aangeschakeld worden. Studenten vergelijken een controle en een behandelde situatie om verschillen te identificeren en de biologische processen achter de response te identificeren. Tegenwoordig is het eenvoudig en goedkoop om het genoom van microbe (bacterie / schimmel) compleet te sequencen. Door verschillende genomen met elkaar te vergelijken kunnen we identificeren welke genen ten grondslag liggen aan bepaalde eigenschappen en evolutionaire veranderingen in biologische processen reconstrueren. Met Next Generation Sequencing technieken kunnen we complexe microbiële ecosystemen in kaart brengen en interacties tussen microben en hun omgeving blootleggen.

Het project:
Studenten bewerken zelf een dataset behorend bij een biologische vraagstelling en presenteren hun resultaten. In deze cursus zal ook aandacht besteed worden aan data management en ethisch aspecten van Big Data. Hoe gaan wetenschappers om met data uit publieke projecten?

Program/Minor

Bachelor biologie; pre-Master's programme Pre-master Biological Sciences

Learning Outcomes

De student leert complexe biologische systemen te begrijpen gebruikmakend van Datasets uit Ecologie, Microbiologie en Moleculaire Biologie; De student leert bewerking en analyse van metagenomics, genomics en RNAseq data met eenvoudige software of zelf geschreven R-scripts; De student leert het principe van lineaire regressie en kan dit toepassen met behulp van R-scripts in de analyse van Ecologische data; De student verkrijgt kennis van publieke Databases waar informatie kan worden gevonden en hoe Datasets kunnen worden opgehaald of opgeslagen; De student begrijpt welke experimentele aanpak er bij Big Data experimenten vereist is; De student kan experimentele gegevens uit databases interpreteren en conclusies trekken over de biologische systemen.

Prerequisites

De studiestof van de vakken uit jaar 1 van de BSc Biologie wordt bekend verondersteld, met name van de vakken Data Verzamelen en Analyseren, Genetica en Evolutie, Microben en Biologie van Planten. Studenten moeten basiskennis hebben van moleculaire biologie, biochemie, microbiologie, plantenbiologie en statistiek in R.

Details

Institution: UvA

Period: Period 4

Level: Intermediate

EC: 6 ec

Life Science Technology and Society

Maatschappelijke context en implicaties van biotechnologische ontwikkelingen (waaronder biofarmaceutische technologie)

Program/Minor

BSc Life Science and Technology

Learning Outcomes

Identificeren van maatschappelijke en ethische aspecten van biotechnologische ontwikkelingen; Identificeren en deconstrueren van normatieve posities van actoren in een maatschappelijk debat; Formuleren van eigen positie ten aanzien van de wenselijkheid van biotechnologische ontwikkelingen met verwijzing naar morele waarden; Je eigen positie effectief communiceren; Mondelinge presentatievaardigheden ontwikkelen; De student kan in een mondelinge presentatie vakinhoudelijke informatie (technisch/academisch) afstemmen op de doelgroep; De student kan de structuurelementen die in de cursus worden aangeboden, toepassen in zijn presentatie; De student kan visualisatiemiddelen ontwerpen die de inhoud van de presentatie ondersteunen en deze bedienen tijdens de presentatie; De student kan stem en lichaamstaal gebruiken om de inhoud van de presentatie te ondersteunen; De student kan professioneel omgaan met vragen uit de zaal

Details

Institution: TU Delft

Period: Period 2

Level: Introductory

EC: 4 ec

Life Sciences

De combinatie van de drie vakken in Q1 van BSc LST: Moleculaire Celbiologie, Life Sciences en Biochemie, geeft een complete introductie van de werking van de levensprocessen in de cel.

Het klonen van DNA voor eiwit expressie en vervolgens zuiveren van een enzym, receptor, biomarker of ander eiwit zijn vaak de eerste stappen die nodig zijn in een nieuw project in de Life Sciences (farmaceutische en biotechnologische industrie). Bij het vak Life Sciences leer je de laboratorium basisvaardigheden die hiervoor nodig zijn: (1) het moleculair klonen, en (2) zuiveren en karakteriseren van eiwitten.

Na voltooien van dit vak kun je de bevindingen van plasmide- en eiwitgerelateerde experimenten, en de karakterisatie van de resultaten, zelfstandig analyseren en deze op basaal, wetenschappelijk niveau, schriftelijk vastleggen. Bovendien kun je op basisniveau een experimenteel protocol ontwerpen of het belang en de rol van de stappen in een aangereikt protocol beschrijven. Ten slotte kun je op basisniveau het doel van een wetenschappelijk artikel in de Moleculaire Celbiologie, Life Sciences en Biochemie begrijplijk samenvatten.

Het vak is gericht om humane ziektes

Program/Minor

BSc Life Science and Technology

Learning Outcomes

De student kan aangereikte biochemische experimenten voorbereiden, en in samenwerking met een partner, uitvoeren en interpreteren; De student kan de stof behandeld in de colleges 'Moleculaire Biologie' en 'Biochemie' toepassen bij de interpretatie van experimentele resultaten; De student kan de uitvoering, resultaten en interpretatie van onderzoek schriftelijk vastleggen op academisch niveau; De student kan begrippen uit moleculaire biologie en biochemie begrijpelijk beschrijven in een beknopt verslag; Transferrable skills: Researching, Independent learning, analysing, written communication, working together

Prerequisites

Middelbare schoolkennis over DNA, eiwitten, en scheikunde

Details

Institution: TU Delft

Period: Period 1

Level: Introductory

EC: 5 ec

Machine Learning (TU Delft)

The goal of the course is to acquaint students with the basic Machine Learning concepts and algorithms. Specifically, the course will cover parametric and non-parametric density estimation, linear and non-linear classification, unsupervised learning including clustering and dimensionality reduction, performance evaluation of predictive algorithms and ethical issues in machine learning.

Program/Minor

BSc Computer Science and Engineering; HBO Technische Informatica - Bridging Programme MSc CS; WO Bridging Programme MSc CS for TUD BSc EE; WO Bridging Programme MSc DSAIT for TUD BSc EE; WO Bridging Programme MSc DSAIT for TUD BSc TW; WO Bridging Programme MSc DSAIT for TUD BSc NB and LST; HBO Technische Informatica - Bridging Programme MSc DSAIT; WO Bridging Programme MSc CS for TUD BSc TW

Learning Outcomes

Explain the basic concepts, mathematics and algorithms of machine learning and their underlying statistical concepts; Implement, apply and evaluate basic ML algorithms in Python; Explain the concept of and identify (implicit) bias in data and ML algorithms.

Prerequisites

Introduction to Programming; Algorithms and Data Structures; Calculus; Linear Algebra; Probability Theory and Statistics

Details

Institution: TU Delft

Period: Period 1

Level: Intermediate

EC: 5 ec

Machine Learning (WUR)

Machine Learning deals with algorithms that predict certain outputs (such as crop yields or traits) given previously unseen input data from cameras, other sensors, maps, molecular measurements etc. These algorithms learn how to do so using training data (sets of input examples, usually with corresponding outputs). Machine learning plays an increasingly important role in many scientific areas.

This course discusses the theory of different methods for regression, classification, and clustering and the application thereof in different fields of agricultural and life sciences. Students will learn how to properly train and evaluate machine-learning models on data, what typical issues are that can arise, and how to deal with these. Furthermore, attention is paid to the ethical, legal, and social aspects of applying machine learning in practical use-cases.

During the course, every day is dedicated to a specific topic, with a lecture and pen-and-paper exercises in the morning, and a practical session with computer exercises in the afternoon. In addition, there are four project days, where students work in pairs to create a solution for practical use-cases.

Program/Minor

Bachelor Data Science for Global Challenges; Master Bioinformatics; Master Biosystems Engineering; Master Data Science for Food and Health; Master Earth and Environment; Master Geo-Information Science

Learning Outcomes

Explain machine-learning problems, algorithms, and their formulas; Qualitatively and quantitatively compare the characteristics, (dis)advantages, formulas, and performance of a number of key algorithms; Evaluate a use-case to choose the right machine-learning method and to properly analyse the results; Apply machine-learning techniques in a) biosystems engineering (MBE), b) bioinformatics (MBF), c) geo-information sciences and remote sensing (MGI), or another field of study; Design and implement effective solutions based on chosen algorithms, to solve practical problems

Prerequisites

Mathematics: Mathematics 1 and Mathematics 2, or equivalent; Statistics: Data Analysis Biosystems Engineering, Advanced Statistics, or equivalent; Programming in Python. There is a significant overlap with the course Statistics for Data Scientists. We recommend students to not take both courses.

Details

Institution: WUR

Period: Period 4

Level: Advanced

EC: 6 ec

Machine Learning 1

This course is lecture based, with homework assignments and programming assignments.

The curriculum is based on chapters 1,2,3,4,5,6,7,9,14 of the book Pattern Recognition and Machine Learning by C. Bishop: Statistical learning principles; Linear regression; Linear classification; Neural networks; Kernel methods; Dimensionality reduction; Clustering methods; Ensemble methods

Program/Minor

MA artificial intelligence & MA Computational Science

Learning Outcomes

The student is able to explain the core concepts and scope of the main areas within machine learning, providing illustrative examples for each. (C2, LTO LR); The student is able to distinguish between the main areas of machine learning (e.g., supervised, unsupervised, deterministic, probabilsitic) by comparing their goals, methods, and application examples. (C4, LTO LR); The student is able to critically evaluate major statistical learning frameworks and principles, analyzing their respective advantages and shortcomings in different contexts. (C5, LTO LR); The student is able to explain the structure, key assumptions, and mathematical formulation of major linear and non-linear statistical models covered in the course. (C2, LTO LR); The student is able to analyze the behaviour and properties of major linear and non-linear statistical models, reasoning about their applicability based on data characteristics. (C4, LTO LR); The student is able to **formulate** appropriate machine learning problems and define learning objectives solvable with the statistical models taught in the course. (C4, LTO LR); The student is able to apply learned machine learning principles and models to analyze real-world datasets and address practical problems using Python. (C3, LTO Skills)

Prerequisites

Calculus, Linear algebra, probability theory, statistics, programming.

Details

Institution: UvA

Period: Period 4

Level: Master

EC: 6 ec

Machine Learning and Introduction to AI

This course teaches the basics of machine learning, including theory and practical aspects, and an introduction to artificial intelligence (AI).

Program/Minor

Electrical Engineering, Mathematics and Computer Science

Learning Outcomes

Discuss about the history of AI, AI definitions, AI approaches, and philosophical arguments regarding AI Apply reflex-, search- and logical-based AI techniques to solve puzzles and games with pen and paper; Apply common operations (pre-processing, scaling, handling missing values) to datasets using Python; Apply the statistical concepts of supervised and unsupervised learning, and recognize their limitations; Analyze which factors impact the performance of learning algorithms; Build and optimize a machine learning pipeline using Python and Scikit-learn and evaluate its performance; Calculate and discuss fairness measures of machine learning models;

Prerequisites

Some statistics and linear algebra is necessary. It is your responsibility to check if you have sufficient prior knowledge. The prior knowledge can be acquired via self-study (15-20h) during the summer. The detailed prior knowledge is also given in the catalogue so you can quickly check if you know all the required concepts.

Details

Institution: TU Delft

Period: Period 1

Level: Advanced

EC: 5 ec

Machine Learning in Bioinformatics

Learning from patterns in molecular biology data plays an important role in diagnosing disease, discovering new targets for therapy, and more generally in answering biological questions that lead to an improved understanding of biological systems with relevance to human health, industry, biotechnology, and agriculture. This course focuses on methodology for the analysis of high-dimensional data in molecular biology, naturally addressing challenges that arise in the field such as learning from unlabelled data or from small numbers of samples. The methodology is introduced in the context of real-world applications with examples using real data.

Covered topics will include (a selection of) the following: Mixture models: (in)finite mixture models, EM algorithm, ECDF, bootstrapping; Clustering: k-means/medoids, density-based clustering, hierarchical clustering, clustering evaluation, choosing number of clusters; Statistical hypothesis testing: p-values, confidence intervals, multiple hypothesis testing, family-wise error rate, (local) false discovery rate; Testing using high-throughput count data: multifactorial designs, (generalized) linear models, analysis of variance, shrinkage estimation; Linear and non-linear dimensionality reduction: matrix decomposition (SVD), biplot representations, PCA, NMF, UMAP, autoencoders; Multivariate methods for heterogenous data: embeddings, multidimensional scaling (MDS), robust (non-metric) MDS, correspondence analysis, canonical correlation; Supervised learning: algorithms, performance metrics, curse of dimensionality, generalizability and model complexity, regularization, cross-validation, ML architectures for sequential data; Interpretability and explainability; Other selected advanced topics, including transfer learning self-supervised learning, semi-supervised learning, multimodal ML.

Program/Minor

MSc Data Science and Artificial Intelligence Technology

Learning Outcomes

Analyse different kinds of high-throughput bioinformatics data and their statistical properties; Discuss published statistical methods and ML algorithms, and the data-driven bioinformatics problems they are designed to solve; Formulate research questions or data-driven problems related to real-world challenges in the field of bioinformatics; Design a scientifically well-reasoned data-driven solution to address problems / research questions in bioinformatics; Evaluate the technical quality and scientific value of data-driven bioinformatics solutions; Communicate on a complete data-driven bioinformatics research project via a scientific report;

Prerequisites

Recommended for fundamental knowledge: CS4220 Machine Learning 1; CS4070 Multivariate Data Analysis; CS4250 Selected Topics in Molecular Biology Required: CSE1200/TI1106 ecM Calculus (or similar) CSE1205/TI1206 ecM Linear Algebra (or similar) CSE1210/TI2216 ecM Probability Theory and Statistics (or similar) The project requires basic programming skills.

Details

Institution: TU Delft

Period: Period 4

Level: Master

EC: 5 ec