Dynamic Workflow Composition: Using Markov Decision Processes.
Prashant DoshiRichard GoodwinRama AkkirajuKunal VermaPublished in: Int. J. Web Serv. Res. (2005)
Keyphrases
- markov decision processes
- state space
- optimal policy
- finite state
- transition matrices
- reinforcement learning
- dynamic programming
- finite horizon
- decision theoretic planning
- policy iteration
- reachability analysis
- factored mdps
- markov decision process
- planning under uncertainty
- partially observable
- decision processes
- infinite horizon
- state and action spaces
- average reward
- dynamic environments
- reinforcement learning algorithms
- model based reinforcement learning
- semi markov decision processes
- risk sensitive
- action space
- average cost
- policy evaluation
- learning algorithm
- reward function
- markov chain
- stochastic shortest path