Importance sampling in reinforcement learning with an estimated behavior policy.
Josiah P. HannaScott NiekumPeter StonePublished in: Mach. Learn. (2021)
Keyphrases
- importance sampling
- reinforcement learning
- monte carlo
- optimal policy
- kalman filter
- markov chain
- particle filter
- particle filtering
- variance reduction
- rare events
- state space
- learning algorithm
- approximate inference
- markov chain monte carlo
- object tracking
- markov decision processes
- image segmentation
- machine learning