Teaching Machine Learning at Aalto University
Machine Learning (ML) is reshaping how we live, work, and innovate. At Aalto University, our group develops and teaches ML with a focus on clarity, accessibility, and critical thinking. We believe that by breaking down ML into simple, universal components, students from all backgrounds can develop both intuition and technical skills.
The Three Fundamental Components of ML
Any ML method can be decomposed into three core components:
- Data Representation – How input data is organized and structured.
- Model Design – The architecture or algorithm that processes the data.
- Loss Function – The criterion used to evaluate and optimize model performance.
This framework helps students understand ML principles without being overwhelmed by unnecessary mathematical formalism.
See related entries in the Aalto Dictionary of Machine Learning
Evidence-Based Teaching Approach
Our courses emphasize active learning and conceptual clarity through:
- Interactive Quizzes and Coding Exercises – reinforcing core concepts.
- Project-Based Assessments – applying ML to real-world problems.
- Peer Review Training – learning to give and receive constructive feedback, the cornerstone of science.
- Learning by Doing – hands-on assignments using Python notebooks.
Teaching Philosophy
- Concise Lectures – focused on clear learning objectives.
- Curated Resources – including the textbook Machine Learning: The Basics by Alex Jung, our YouTube channel, and GitHub repositories.
- Feedback-Adaptive Cycle – course materials are regularly revised based on student feedback.
Recognition and Impact
-
2018 – Teacher of the Year, Department of Computer Science, Aalto University.
Teacher of the Year award certificate -
2020 – Teaching Assistant of the Year, awarded to Shamsi Abdurakhmanova, Aalto School of Science.
Read announcement on Aalto.fi -
2020–2024 – Most popular course in FiTech:
CS-EJ3211 Machine Learning with Python, the largest registered course across the Finnish network university, with outstanding student feedback.
Read more (PDF) - ⭐ High Student Ratings – consistently excellent teaching evaluations.
- 🌍 International Engagement – invited speaker and lecturer at institutions and conferences worldwide.
Our Approach in a Nutshell
We combine conceptual clarity (data, models, loss), active learning (projects, peer review, coding exercises), and constant adaptation (feedback-driven improvements). Our work has been recognized both in Finland and internationally, and we continue to make our teaching openly available whenever possible.
Keywords
machine learning teaching, data representation, model design, loss functions, interactive learning, project-based assessments, Python notebooks, JupyterHub, Aalto University, teacher of the year, international lecturer