Scalable Safety-Critical Policy Evaluation with Accelerated Rare Event Sampling.
Mengdi XuPeide HuangFengpei LiJiacheng ZhuXuewei QiKentaro OguchiZhiyuan HuangHenry LamDing ZhaoPublished in: IROS (2022)
Keyphrases
- policy evaluation
- importance sampling
- safety critical
- monte carlo
- rare events
- variance reduction
- temporal difference
- formal methods
- agent architecture
- embedded systems
- fault tolerant
- least squares
- reinforcement learning
- markov chain
- kalman filter
- markov chain monte carlo
- particle filter
- model free
- machine learning
- sample size
- markov decision processes
- test set
- minority class
- multi agent systems