On Online Learning in Kernelized Markov Decision Processes.
Sayak Ray ChowdhuryAditya GopalanPublished in: CoRR (2019)
Keyphrases
- markov decision processes
- online learning
- state space
- reinforcement learning
- optimal policy
- transition matrices
- finite state
- dynamic programming
- policy iteration
- e learning
- markov decision process
- kernel function
- reachability analysis
- decision theoretic planning
- active learning
- average cost
- finite horizon
- decision processes
- planning under uncertainty
- risk sensitive
- action sets
- partially observable
- reinforcement learning algorithms
- action space
- average reward
- reward function
- infinite horizon
- model based reinforcement learning
- semi markov decision processes
- data mining
- discounted reward
- state and action spaces
- collaborative filtering