Safe Model-Based Multi-Agent Mean-Field Reinforcement Learning.
Matej JusupBarna PásztorTadeusz JanikKenan ZhangFrancesco CormanAndreas KrauseIlija BogunovicPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- multi agent
- model free
- multi agent systems
- function approximation
- multi agent reinforcement learning
- reinforcement learning algorithms
- cooperative
- multi agent environments
- statistical mechanics
- single agent
- temporal difference
- autonomous agents
- intelligent agents
- action selection
- belief networks
- multiagent systems
- markov random field
- reinforcement learning agents
- function approximators
- learning algorithm
- traffic signal control
- markov networks
- closed form
- optimal policy
- state space
- dynamic programming
- optimal control
- linear complexity
- fully unsupervised
- heterogeneous agents
- state abstraction
- multiagent reinforcement learning
- learning process
- neural network