On the Complexity of Reachability in Parametric Markov Decision Processes.
Tobias WinklerSebastian JungesGuillermo A. PérezJoost-Pieter KatoenPublished in: CONCUR (2019)
Keyphrases
- markov decision processes
- state space
- optimal policy
- finite state
- reinforcement learning
- policy iteration
- transition matrices
- dynamic programming
- decision theoretic planning
- partially observable
- action space
- decision problems
- action sets
- reinforcement learning algorithms
- planning under uncertainty
- reachability analysis
- factored mdps
- risk sensitive
- decision processes
- computational complexity
- heuristic search
- markov chain
- finite horizon
- average cost
- state abstraction
- infinite horizon
- model based reinforcement learning
- average reward
- policy evaluation
- markov decision process
- partially observable markov decision processes
- reward function
- planning problems