This repository is for instantiating agents to conpare their competency in Tic-Tac-Toe.
- Agent using DQN algorythm
- Agent using Minimax algorythm
- Agent using intelligent but not complete algorythm
Try to experiment it and play with the agent by
python dqn.py
or
docker build -t dqnpy . && docker run --rm -i dqnpy
The program stops if the number of learning you input is big, but start after a while.