Minimax Regret Optimisation for Robust Planning in Uncertain Markov Decision Processes.
Marc RigterBruno LacerdaNick HawesPublished in: CoRR (2020)
Keyphrases
- markov decision processes
- minimax regret
- planning under uncertainty
- decision theoretic planning
- reward function
- partially observable
- state space
- reinforcement learning
- optimal policy
- transition matrices
- dynamic programming
- policy iteration
- infinite horizon
- heuristic search
- partially observable markov decision processes
- markov decision process
- utility function
- incomplete information
- planning problems
- preference elicitation
- markov decision problems
- probabilistic planning
- multi agent
- decision making