Skip to content
Projekte

mlii/mfrl has 424 GitHub stars and 106 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-19 17:42 UTC. Last push 2020-03-11.

So werden diese Zahlen gemessen

Gemessener Sternzuwachs

in 24h
0
in 7 days
Verlauf wird gesammelt
in 15 days
Verlauf wird gesammelt
in 1 month
Verlauf wird gesammelt

Nettoänderungen in überlappenden Zeiträumen, keine Tagessummen. Fehlende Messwerte bleiben leer. Sterne zeigen Interesse, nicht Qualität.

Über uns

Mean Field Multi-Agent Reinforcement Learning

Lernen

README.md

Original

Mean Field Multi-Agent Reinforcement Learning

Implementation of MF-Q and MF-AC in the paper Mean Field Multi-Agent Reinforcement Learning .

Example

image

An 20x20 Ising model example under the low temperature.

A 40x40 Battle Game gridworld example with 128 agents, the blue one is MFQ, and the red one is IL.

Code structure

  • main_MFQ_Ising.py: contains code for running tabular based MFQ for Ising model.

  • ./examples/: contains scenarios for Ising Model and Battle Game (also models).

  • battle.py: contains code for running Battle Game with trained model

  • train_battle.py: contains code for training Battle Game models

Compile Ising environment and run

Requirements

  • python==3.6.1
  • gym==0.9.2 (might work with later versions)
  • matplotlib if you would like to produce Ising model figures

Compile MAgent platform and run

Before running Battle Game environment, you need to compile it. You can get more helps from: MAgent

Steps for compiling

cd examples/battle_model
./build.sh

Steps for training models under Battle Game settings

  1. Add python path in your ~/.bashrc or ~/.zshrc:

    vim ~/.zshrc
    export PYTHONPATH=./examples/battle_model/python:${PYTHONPATH}
    source ~/.zshrc
    
  2. Run training script for training (e.g. mfac):

    python3 train_battle.py --algo mfac
    

    or get help:

    python3 train_battle.py --help
    

Paper citation

If you found it helpful, consider citing the following paper:




@InProceedings{pmlr-v80-yang18d,
  title = 	 {Mean Field Multi-Agent Reinforcement Learning},
  author = 	 {Yang, Yaodong and Luo, Rui and Li, Minne and Zhou, Ming and Zhang, Weinan and Wang, Jun},
  booktitle = 	 {Proceedings of the 35th International Conference on Machine Learning},
  pages = 	 {5567--5576},
  year = 	 {2018},
  editor = 	 {Dy, Jennifer and Krause, Andreas},
  volume = 	 {80},
  series = 	 {Proceedings of Machine Learning Research},
  address = 	 {Stockholmsmässan, Stockholm Sweden},
  month = 	 {10--15 Jul},
  publisher = 	 {PMLR}
}