The Complexity of Graph-Based Reductions for Reachability in Markov Decision Processes.
Stéphane Le RouxGuillermo A. PérezPublished in: FoSSaCS (2018)
Keyphrases
- markov decision processes
- state space
- optimal policy
- dynamic programming
- finite state
- transition matrices
- reachability analysis
- decision theoretic planning
- policy iteration
- factored mdps
- decision problems
- reinforcement learning algorithms
- reinforcement learning
- average reward
- action space
- partially observable
- decision processes
- planning under uncertainty
- infinite horizon
- risk sensitive
- finite horizon
- heuristic search
- action sets
- average cost
- markov chain
- partially observable markov decision processes
- reward function
- state and action spaces
- linear programming
- semi supervised
- semi markov decision processes
- computational complexity