Incremental Value Iteration for Time-Aggregated Markov-Decision Processes.
Tao SunQianchuan ZhaoPeter B. LuhPublished in: IEEE Trans. Autom. Control. (2007)
Keyphrases
- markov decision processes
- state space
- finite state
- reinforcement learning
- policy iteration
- optimal policy
- decision theoretic planning
- transition matrices
- partially observable
- factored mdps
- infinite horizon
- stochastic shortest path
- reachability analysis
- average cost
- average reward
- dynamic programming
- planning under uncertainty
- reinforcement learning algorithms
- risk sensitive
- finite horizon
- markov decision process
- semi markov decision processes
- model based reinforcement learning
- action space
- discounted reward
- reward function
- decision processes
- incremental algorithms
- action sets
- multistage