On existence of Berk-Nash equilibria in misspecified Markov decision processes with infinite spaces.
Robert M. AndersonHaosui DuanmuAniruddha GhoshM. Ali KhanPublished in: J. Econ. Theory (2024)
Keyphrases
- markov decision processes
- nash equilibria
- stochastic games
- congestion games
- incomplete information
- finite state
- stationary policies
- game theory
- state space
- optimal policy
- average reward
- game theoretic
- dynamic programming
- reinforcement learning
- pure strategy
- nash equilibrium
- policy iteration
- reinforcement learning algorithms
- partially observable
- average cost
- finite horizon
- infinite horizon
- decision theoretic planning
- transition matrices
- markov decision process
- reward function
- cost function
- cooperative
- first order logic
- mixed strategy
- pure nash equilibria