Finite-horizon Equilibria for Neuro-symbolic Concurrent Stochastic Games.
Rui YanGabriel SantosXiaoming DuanDavid ParkerMarta KwiatkowskaPublished in: CoRR (2022)
Keyphrases
- stochastic games
- finite horizon
- games with incomplete information
- markov decision processes
- infinite horizon
- nash equilibria
- repeated games
- long run
- optimal policy
- average reward
- nash equilibrium
- incomplete information
- reinforcement learning
- dynamic programming
- average cost
- game theory
- state space
- markov decision process
- finite state
- reinforcement learning algorithms
- multistage
- policy iteration
- neural network
- game theoretic
- non stationary
- lead time
- decision making