On the Maximization of Long-Run Reward CVaR for Markov Decision Processes.
Li XiaZhihui YuPeter W. GlynnPublished in: CoRR (2023)
Keyphrases
- long run
- markov decision processes
- optimal policy
- average reward
- short run
- average cost
- reinforcement learning
- reward function
- discounted reward
- infinite horizon
- finite state
- total reward
- dynamic programming
- decision problems
- state space
- policy iteration
- robust optimization
- finite horizon
- expected cost
- expected reward
- markov decision process
- reinforcement learning algorithms
- objective function
- transition matrices
- queueing networks
- sufficient conditions
- decision theoretic planning
- state and action spaces
- multistage
- risk sensitive
- partially observable
- control policy
- action space
- exchange rate
- least squares
- data mining