Stopping Criteria for Value Iteration on Stochastic Games with Quantitative Objectives.
Jan KretínskýTobias MeggendorferMaximilian WeiningerPublished in: LICS (2023)
Keyphrases
- stochastic games
- stopping criteria
- markov decision processes
- average reward
- infinite horizon
- state space
- reinforcement learning algorithms
- reinforcement learning
- policy iteration
- nash equilibria
- clustering algorithm
- dynamic programming
- optimal policy
- long run
- finite state
- repeated games
- nash equilibrium
- markov chain
- imperfect information
- multi agent
- heuristic search
- learning automata
- search algorithm
- markov decision process
- partially observable markov decision processes
- single agent
- partially observable
- stopping criterion
- genetic algorithm