Approximately Solving Mean Field Games via Entropy-Regularized Deep Reinforcement Learning.
Kai CuiHeinz KoepplPublished in: AISTATS (2021)
Keyphrases
- reinforcement learning
- reinforcement learning agents
- learning agents
- function approximation
- markov random field
- action sets
- state space
- information theoretic
- game theoretic
- objective function
- multiagent learning
- markov decision problems
- imperfect information
- belief networks
- multi agent
- least squares
- information theory
- reinforcement learning algorithms
- multi agent systems
- statistical mechanics
- learning process
- partition function
- mutual information
- em algorithm
- nash equilibria
- temporal difference
- model free
- solving problems
- markov decision processes
- game playing
- educational games
- optimal control
- nash equilibrium
- game theory