david-spc / Nash-DQN

Deep Reinforcement Learning for Nash Equilibria

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

This repository contains the code for the Nash-DQN algorithm for general-sum multi-agent reinforcement learning. The associated paper "Deep Q-Learning for Nash Equilibria: Nash-DQN" can be found at https://arxiv.org/abs/1904.10554.

INSTRUCTIONS:

To generate plots based on pre-trained network:

  • Open file "Visualization.ipynb" and run all cells

To train network based on default parameters:

  • Open file "Training.ipynb" and run cell with appropiate parameters
  • Open file "Visualization.ipynb"
  • Change variable net_file_name to the designated file name of Action Network
  • Run all cells to generate plots

About

Deep Reinforcement Learning for Nash Equilibria


Languages

Language:Jupyter Notebook 85.2%Language:Python 14.8%