Applied Machine Learning

Course manual 2025/2026

Course content

Machine learning has moved from a niche research field to a core technology underpinning modern industry and research. Applications now span spam filtering, web search, personalized recommendations, medical imaging, weather modeling, protein folding, and more. This course introduces the principles and standard practices of applied machine learning, with an emphasis on both conceptual understanding and hands-on experience.

We will study the complete machine learning pipeline: problem formulation, data preprocessing, model training, and evaluation. The course begins with some foundations in classification, regression, matrix arithmetic, and calculus before moving onto fundamental approaches in deep learning. We have application lectures on topics such as text processing, computer vision, reinforcement learning, and recommender systems.

The course combines lectures, labs, and math tutorials. In the lectures, you will learn about core theoretical concepts in machine learning. In the labs, you will gain hands-on experience via coding assignments and by participating in a Kaggle competition for your group project. In the math tutorials, we will focus on some off the important mathematical concepts needed to understand how machine learning systems work. We will also include some guest lectures from both academic and industry experts provide additional perspective on current challenges and applications of machine learning in the real world.

Study materials

Syllabus

• A syllabus containing articles and chapters will be made available at the beginning of the course. 

   

  The syllabus/course will cover parts of Deep Learning, by Ian Goodfellow, Yoshua Bengio, and Aaron Courville MIT Press, 2016, see http://www.deeplearningbook.org 

Objectives

• Identify standard machine learning terminology, problem types, and core algorithms (e.g., classification, regression, neural networks).
• Explain how data characteristics, model architecture, and optimization procedures affect model performance.
• Compute core mathematical calculations used in deep learning systems (e.g., vector/matrix operations, gradient calculations).
• Implement machine learning models and experiments by writing code independently in Python using standard libraries (e.g., numpy, pandas, PyTorch, matplotlib).
• Critically assess the social and ethical implications of deploying machine learning models in the real world, and justify a particular position.
• Design, implement and present a machine learning system that is optimized for a given task, including model selection rationale and empirical validation.

Teaching methods

• Lecture
• Self-study
• Presentation/symposium
• Laptop seminar
• Working independently on e.g. a project or thesis

Lectures will the cover core content of the course. 

Laptop seminars will provide students with the opportunity to ask TAs for help with the assignments and group project. 

Presentation is for the group project, where students will need to communicate their results in a coherent and comprehensive manner. 

Self-study is essential for ensuring complete comprehension of the course material. 

Group project is to practice implementing your own ML solution. 

Learning activities

Attendance

In TER part B of this programme no requirements regarding attendance are mentioned.

Additional requirements for this course:

Although attendance is not taken formally, students are expected to attend all lectures. 

Assessment

The final grade consists of (1) a midterm exam, (2) a final exam, and (3) a group project. A minimum grade of 5.5/10 is required for both the final exam and the group project in order to pass the course.

All deadlines are strict. Late submissions incur a penalty of 0.5 points (out of 10) per day. Exceptions are granted only in exceptional circumstances and at the discretion of the course coordinator.

The midterm and final exams assess material covered in lectures, including the mathematics sessions. Both exams are handwritten.

Only the final exam has a resit. If the resit is taken, the resit grade replaces the original final exam grade and is capped at 5.5/10. If the midterm exam is missed, its weight is transferred to the final exam. There will be no exceptions to this grading scheme.

Assignments

Midterm exam

• The midterm exam covers all material addressed in the lectures. It is individual and graded. 

Final exam

• The exam covers all material addressed in the lectures. It is individual and graded. 

Group Project

• The project is about implementing a solution for a Kaggle challenge. It is done in groups and is graded. 

Quizzes

• The quizzes cover all material addressed in the lectures. They are meant as preparation for the midterm and exam. They are individual and ungraded. 

Coding assignments

• The coding assignments cover material addressed in the lectures are tutorials. They are meant as preparation for the midterm and exam. They are individual and ungraded. 

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Course structure

Additional information

Note: MSc AI students are not allowed to take this course due to the significant overlap with Machine Learning 1, Computer Vision 1 and  Deep Learning 1.

Contact information

Coordinator

• dr. A. Lucic