Semi-Infinitely Constrained Markov Decision Processes and Provably Efficient Reinforcement Learning.
Liangyu ZhangYang PengWenhao YangZhihua ZhangPublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2024)
Keyphrases
- markov decision processes
- reinforcement learning
- optimal policy
- state space
- policy iteration
- finite state
- dynamic programming
- reinforcement learning algorithms
- partially observable
- transition matrices
- action space
- finite horizon
- average reward
- decision theoretic planning
- state and action spaces
- state abstraction
- risk sensitive
- planning under uncertainty
- model based reinforcement learning
- markov decision process
- infinite horizon
- factored mdps
- reachability analysis
- action sets
- average cost
- stochastic games
- policy evaluation
- total reward
- discounted reward
- markov decision problems
- least squares
- multistage
- function approximation
- finite number
- model free
- semi markov decision processes
- multi agent
- continuous state spaces
- search space
- reward function