MARL Cooperative Path Planning

Forthcoming changes:

Code refactor with docstrings
Getting started / tutorial guide
Documented jupyter notebooks
Implementation of probabilistic agents (using Pyro).
Improved meta-protocol design (inc. aggregation function) and message passing capabilities.

Based on: MSc Thesis on Multi-Agent Reinforcement Learning for Path Planning (applied to PCB routing)[2020]

Name		Name	Last commit message	Last commit date
Latest commit History 36 Commits
__pycache__		__pycache__
agents		agents
pcb_twin		pcb_twin
.DS_Store		.DS_Store
Conv MLP DQN Prioritised .ipynb		Conv MLP DQN Prioritised .ipynb
DRQN with Episodic Replay .ipynb		DRQN with Episodic Replay .ipynb
MARL DQN Prioritised.ipynb		MARL DQN Prioritised.ipynb
MARL DRQN Distributed Vanilla.ipynb		MARL DRQN Distributed Vanilla.ipynb
MARL DRQN RIAL.ipynb		MARL DRQN RIAL.ipynb
README.md		README.md
learning_history.csv		learning_history.csv
replay_buffers.py		replay_buffers.py
thesis_img.png		thesis_img.png

Provide feedback