Multiagent Reinforcement Learning toolbox Version 1.2 This directory contains the main control functions of the toolbox. learn - Implements the learning control mechanism replay - Replays the agents' current policies episodic_learn - Implements the learning control mechanism in an episodic environment episodic_replay - Replays the agents' current policies in an episodic environment startuprl - Startup script for MA-RL toolbox cleanuprl - Cleans up the paths added by the MA-RL startup script Changelog - A changelog for the MA-RL toolbox Subdirectories: /agent - Contains the functions and templates for implementing the agent behaviour /world - Contains the functions and templates for implementing worlds (tasks) /batch - Contains utility functions for running and processing batches of experiments /lib - Contains library functions required by the MARL toolbox /rldemo - Contains demonstrations of the usage and features of the MARL toolbox