Differentially Private Regret Minimization in Episodic Markov Decision Processes.
Sayak Ray ChowdhuryXingyu ZhouPublished in: AAAI (2022)
Keyphrases
- markov decision processes
- differentially private
- regret minimization
- game theoretic
- differential privacy
- nash equilibrium
- optimal policy
- state space
- transition matrices
- reinforcement learning
- policy iteration
- finite state
- dynamic programming
- action space
- reinforcement learning algorithms
- infinite horizon
- decision theoretic planning
- markov decision process
- average reward
- multi agent learning
- stochastic games
- reward function
- average cost
- fixed point
- game theory
- search engine