Partial and Conditional Expectations in Markov Decision Processes with Integer Weights.
Jakob PiribauerChristel BaierPublished in: CoRR (2019)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- state space
- optimal policy
- decision theoretic planning
- dynamic programming
- policy iteration
- risk sensitive
- action space
- average reward
- reachability analysis
- partially observable
- transition matrices
- reinforcement learning algorithms
- factored mdps
- state and action spaces
- planning under uncertainty
- decision processes
- reward function
- average cost
- linear combination
- finite horizon
- interval estimation
- stochastic shortest path
- model based reinforcement learning
- data mining
- action sets
- infinite horizon
- search algorithm
- bayesian networks
- decision making
- machine learning