Efficient Learning in Non-Stationary Linear Markov Decision Processes.
Ahmed TouatiPascal VincentPublished in: CoRR (2020)
Keyphrases
- non stationary
- markov decision processes
- efficient learning
- optimal policy
- state space
- finite state
- finite horizon
- reinforcement learning algorithms
- transition matrices
- decision theoretic planning
- model based reinforcement learning
- policy iteration
- reinforcement learning
- dynamic programming
- planning under uncertainty
- average reward
- reachability analysis
- factored mdps
- random fields
- learning algorithm
- infinite horizon
- average cost
- concept drift
- action space
- partially observable
- reward function
- markov decision process
- action sets
- pattern languages
- membership queries
- object oriented
- state and action spaces
- discounted reward