Name		Name	Last commit message	Last commit date
parent directory ..
README.md		README.md
agent.py		agent.py
illustration.svg		illustration.svg
network.py		network.py
nim.py		nim.py
train.ipynb		train.ipynb

README.md

Multi-agent DQN for Nim-21

Two agents that learn to play Nim-21 using PyTorch and DQN. Both agents learn the game-theory optimal strategy.