Coevolutive planning in markov decision processes.
Bruno ScherrerFrançois CharpilletPublished in: AAMAS (2002)
Keyphrases
- markov decision processes
- macro actions
- planning under uncertainty
- partially observable
- decision theoretic planning
- state space
- finite state
- planning problems
- transition matrices
- optimal policy
- reinforcement learning
- policy iteration
- finite horizon
- decision processes
- probabilistic planning
- reachability analysis
- partially observable markov decision processes
- infinite horizon
- heuristic search
- model based reinforcement learning
- dynamic programming
- factored mdps
- average cost
- risk sensitive
- average reward
- reinforcement learning algorithms
- ai planning
- action space
- markov decision process
- path finding
- optimal solution
- stochastic shortest path
- discounted reward
- expected reward
- state and action spaces
- markov decision problems
- classical planning
- goal oriented
- decision theoretic