Balancing Two-Player Stochastic Games with Soft Q-Learning.
Jordi Grau-MoyaFelix LeibfriedHaitham Bou-AmmarPublished in: IJCAI (2018)
Keyphrases
- stochastic games
- reinforcement learning algorithms
- state action
- multi agent reinforcement learning
- reinforcement learning
- single agent
- multi agent
- nash equilibria
- model free
- markov decision processes
- state space
- repeated games
- learning agent
- function approximation
- multiagent reinforcement learning
- temporal difference
- imperfect information
- learning algorithm
- rl algorithms
- nash equilibrium
- multiple agents
- average reward
- reward function
- dynamic environments
- dynamic programming
- cooperative
- multi agent systems
- learning automata
- optimal policy
- game theory
- long run
- decision problems
- game theoretic
- infinite horizon