REValueD: Regularised Ensemble Value-Decomposition for Factorisable Markov Decision Processes.
David IrelandGiovanni MontanaPublished in: CoRR (2024)
Keyphrases
- markov decision processes
- state space
- finite state
- optimal policy
- reinforcement learning
- dynamic programming
- reachability analysis
- policy iteration
- transition matrices
- planning under uncertainty
- partially observable
- decision processes
- average cost
- finite horizon
- action space
- factored mdps
- model based reinforcement learning
- decision theoretic planning
- risk sensitive
- infinite horizon
- average reward
- reinforcement learning algorithms
- markov decision process
- state and action spaces
- action sets
- stochastic shortest path