Extended Markov Games to Learn Multiple Tasks in Multi-Agent Reinforcement Learning
Artificial Intelligence
2020-02-17 v1 Machine Learning
Logic in Computer Science
Multiagent Systems
Abstract
The combination of Formal Methods with Reinforcement Learning (RL) has recently attracted interest as a way for single-agent RL to learn multiple-task specifications. In this paper we extend this convergence to multi-agent settings and formally define Extended Markov Games as a general mathematical model that allows multiple RL agents to concurrently learn various non-Markovian specifications. To introduce this new model we provide formal definitions and proofs as well as empirical tests of RL algorithms running on this framework. Specifically, we use our model to train two different logic-based multi-agent RL algorithms to solve diverse settings of non-Markovian co-safe LTL specifications.
Cite
@article{arxiv.2002.06000,
title = {Extended Markov Games to Learn Multiple Tasks in Multi-Agent Reinforcement Learning},
author = {Borja G. León and Francesco Belardinelli},
journal= {arXiv preprint arXiv:2002.06000},
year = {2020}
}
Comments
Long version of the correspondent ECAI 2020 paper