VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.
Hands-On Deep Learning
Last Updated: 2026-06-03 00:14:20
Abstract
This lab offers hands-on deep learning exercises using PyTorch, covering computer vision, audio processing, graph neural networks, natural language processing, reinforcement learning, and representation learning. The material is organized into six topics, each spanning two weeks.
Objective
The goal of this course is to introduce students to both fundamental and advanced neural network architectures, helping them understand how these models work through practical examples. Students will study different types of networks as building blocks and apply them to solve problems and complete course challenges. By the end of the course, students will be familiar with architectures such as multi-layer perceptrons, convolutional neural networks, recurrent neural networks, transformers, graph-based networks (e.g., graph convolutional and attention networks), and autoencoders. They will also learn how to train a network from scratch or fine-tune a pre-trained model for various tasks and data types.
Resources
Lecture Notes
Python Notebooks will be distributed to students before every session.
Learning Materials (Links)
- Main link
- Website
General Information
- Language
- English
- Levels
- BSC
- Frequency
- Semesterly recurring
Examination
- Type
- ungraded semester performance
Registration & Places
- Max Places
- 200
Course Components
| Type | Title | Time & Place | Hours |
|---|---|---|---|
| practical/laboratory course | Hands-On Deep Learning |
|
32 h semesterly |
Offered In
-
-
Electives (Students may also choose courses from the Master's program in Computer Science. It is their responsibility to make sure that they meet the requirements and conditions for these courses.)
-
-
-
Laboratory Courses, Projects, Seminars (A minimum of 15 cp must be achieved in the category "Laboratory Courses, Projects, Seminars")
-