E6885 Reinforcement Learning Final project: Doom Slayer

Tingyu Mao, Kaili Chen, Jingyu Qian, Tian Zhou, Lingyu Zhang

We want to create an agent which can perform well in Doom, a first-person shooting game.

Our initial goal is to pass the basic test scenarios in ./example/scenarios/ folder.

Ref:

• https://github.com/itaicaspi/keras-dqn-doom
• https://github.com/fhennecker/deepdoom
• Prioritized replay: https://github.com/MorvanZhou/Reinforcement-learning-with-tensorflow
• A3C: https://github.com/awjuliani/DeepRL-Agents