Restarted Bayesian Online Change-point Detection for Non-Stationary Markov Decision Processes.
Réda AlamiMohammed MahfoudEric MoulinesPublished in: CoRR (2023)
Keyphrases
- markov decision processes
- change point detection
- non stationary
- change point
- transition matrices
- optimal policy
- reinforcement learning
- state space
- finite horizon
- finite state
- dynamic programming
- policy iteration
- normalized maximum likelihood
- random fields
- sequential data
- outlier detection
- action sets
- decision theoretic planning
- average reward
- action space
- infinite horizon
- markov decision process
- reward function
- average cost
- sequence data
- gaussian processes
- least squares
- hidden markov models
- multiresolution