Beyond Average Return in Markov Decision Processes.
Alexandre MartheAurélien GarivierClaire VernadePublished in: CoRR (2023)
Keyphrases
- markov decision processes
- average cost
- finite state
- discounted reward
- optimal policy
- reinforcement learning
- policy iteration
- state space
- decision processes
- risk sensitive
- dynamic programming
- transition matrices
- partially observable
- reinforcement learning algorithms
- finite horizon
- factored mdps
- planning under uncertainty
- reachability analysis
- average reward
- initial state
- data mining
- action space
- infinite horizon
- action sets
- semi markov decision processes
- multi agent
- state and action spaces
- linear programming
- total reward
- state abstraction
- long run
- stationary policies
- markov decision process