392263 ISY Project: Artificial Intelligence for Computer Games (Pj) (SoSe 2020)

In the project, you will gain practical experience with Learning Architectures for Artificial Intelligence in simulated environments, apply neural networks, and Human-Brain inspired approaches (eye-movements, visual search, attention-based mechanisms, symbolic reasoning) to learn and produce complex purposeful actions in the context of one of the following environments on your choice [1-8]. Advanced students will get a chance to participate in AI competitions, e.g. [9]

[1] Atari: https://youtu.be/XjsY8-P4WHM?t=48
[2] Flatlan trains: https://www.aicrowd.com/challenges/flatland-challenge
[3] Pommerman: https://www.pommerman.com
[4] Montezuma Revenge: https://gym.openai.com/envs/MontezumaRevenge-v0
[5] Doom: http://vizdoom.cs.put.edu.pl
[6] Character Control (PFNN): https://youtu.be/Ul0Gilv5wvY
[7] Character Control (DeepMimic) https://youtu.be/vppFvq2quQ0
[8] Challenges: https://www.aicrowd.com/challenges
[9] 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).