Welcome

This Jupyter Book explores a range of deep learning topics, gradually progressing from fundamental concepts to more advanced applications. The material was initially developed for the deep learning course in the Brain and Behaviour Master’s programme at Justus-Liebig University of Giessen. While some examples are specifically designed for Experimental Psychologists and Cognitive Neuroscientists, the content is valuable for anyone interested in deep learning.

Prerequisites

• 0. Environment Setup

Tutorials

• 1. Network’s Building Blocks
• 2. Quick Start
• 3. Optimisation and Learning
• 4. Vision
• 5. Deep Generative Models
• 6. Interpretation Techniques
• 7. Big Projects
• 8. Other Modalities
• 9. Reinforcement Learning

Assignments

• 1. Warming-up
• 2. Dataloaders
• 3. Optimisation and Learning

Student projects

• Deep Learning with Dobble

Python course

• Beginners

Play with notebooks!

The best way to learn the presented materials is by getting your hands dirty! All Jupyter Notebooks can be executed on Google Colab by directly clicking on the icon on top of each notebook. All materials are also available in our GitHub repository , clone it to run it locally on your computer.

Please give feedback!

Please report any issues/bugs/typos/etc. you notice. Either go the issue tracker or directly click on the icon on top of each page and then “open issue”. We also accept requests for new materials, and of course, we highly appreciate it if you decide to contribute by pull requests .

Background materials

Python tutorials

While the tutorials explain Python codes in great detail, familiarity with Python is required to follow the course. Useful materials for inexperienced Python programmers:

1. The Python Tutorial

2. W3Schools Python Tutorial

3. Python Numpy Tutorial

Recommended resources

There are several online resources covering deep-learning theoretical and practical aspects:

1. Neural Networks and Deep Learning

2. Dive into Deep Learning

3. UvA Deep Learning Tutorials