Deep Reinforcement Learning for Multi Agent Soccer

TODO

• Try curriculum based approach
• Alternative single controller with some noise.
• Fixing strategy

References:

• Opponent Modeling in Deep Reinforcement Learning He He , Jordan Boyd-Graber, Kevin Kwok , Hal Daume III
• Learning Multiagent Communication with Backpropagation Sainbayar Sukhbaatar, Arthur Szlam, Rob Fergus
• https://github.com/LARG/HFO
• https://github.com/fchollet/keras.git
• https://github.com/matthiasplappert/keras-rl