The Power of Exploiter: Provable Multi-Agent RL in Large State Spaces.
Chi JinQinghua LiuTiancheng YuPublished in: ICML (2022)
Keyphrases
- multi agent
- reinforcement learning
- state space
- reinforcement learning algorithms
- markov decision processes
- cooperative
- markov chain
- learning agents
- optimal policy
- multi agent systems
- power consumption
- action space
- heuristic search
- software agents
- agent oriented
- single agent
- model free
- multiagent reinforcement learning
- autonomous agents
- intelligent agents
- dynamic programming
- planning problems
- temporal difference
- multiagent systems
- markov decision process
- learning algorithm
- multi agent environments
- neural network
- function approximation
- action selection
- dynamical systems
- electricity markets
- pattern databases
- team formation