392264 ISY Project: Deep Reinforcement Learning (Pj) (SoSe 2020)

The project intends to gain practical experience with Deep Reinforcement Learning [1]. You will apply neural network models to learn and produce complex purposeful actions in simulated environments in the context of: robotics [2], locomotion [3] and/or computer games [4, 5, 6]. Advanced students will get a chance to participate in AI competitions, e.g. [7].

[1] Deep Reinforcement Learning: https://deepmind.com/blog/article/deep-reinforcement-learning
[2] OpenAI robotics:
https://blog.openai.com/ingredients-for-robotics-research
[3] Learning to Walk:
https://youtu.be/WuqNdNBVzzI
[4] Doom: http://vizdoom.cs.put.edu.pl
[5] Unity Obstacle Tower Challenge: https://youtu.be/owKdLnCjy3o
[6] Pommerman: https://www.pommerman.com
[7] Microsoft Research AI challenge winners: https://twitter.com/MSFTResearch/status/1099006255857713153
In case this would not find enough interest for a team project, this project proposal would be also offered (in reduced/modified form)
[x] as individual project 
[x] as project for 2-3 students

Teilnahmevoraussetzungen, notwendige Vorkenntnisse

- Introduction to Neural Networks or Advanced Neural Networks courses.
- Python or C++ ( > 1 year).

Literaturangaben

A Free course in Deep Reinforcement Learning from beginner to expert
https://simoninithomas.github.io/Deep_reinforcement_learning_Course