Machine Learning and Computational Complexity

ETH Zürich: Fall 2025

Course Description

This Bachelor’s seminar explores the computational foundations of machine learning. Creating algorithms that learn general rules from data is the backbone of modern artificial intelligence and data-driven discovery; we will explore the theoretical principles that dictate the potential and limitations of learning algorithms. Throughout the seminar, we will investigate several fundamental themes. We’ll examine the crucial trade-off between a model’s complexity and its ability to generalize to new, unseen data. We’ll ask what makes certain learning problems computationally hard while others are easy, and how the quantity, quality, and representation of data impact an algorithm’s performance. We’ll consider how interactive feedback and different learning protocols can lead to more efficient and powerful models.

Time: Friday 14–16h

Location: CHN D 44

For inquiries, please email Francesco Re at francesco.re@inf.ethz.ch.

When you send an email, please always include “Bachelor’s Seminar” in the subject line.

Course Schedule