Approximately Solving Mean Field Games via Entropy-Regularized Deep Reinforcement Learning.
Kai CuiHeinz KoepplPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- reinforcement learning agents
- learning agents
- information theory
- markov random field
- function approximation
- least squares
- temporal difference learning
- statistical mechanics
- information theoretic
- game playing
- video games
- belief networks
- multiagent learning
- linear complexity
- model free
- action sets
- machine learning
- computer games
- mutual information
- state space
- optimal control
- bayesian inference
- game play
- game theory
- action selection
- temporal difference
- first order logic
- optimal policy
- multi agent
- bayesian networks
- learning algorithm