Conservative Bayesian Model-Based Value Expansion for Offline Policy Optimization.
Jihwan JeongXiaoyu WangMichael GimelfarbHyunwoo KimBaher AbdulhaiScott SannerPublished in: CoRR (2022)
Keyphrases
- global optimization
- optimization algorithm
- data driven
- real time
- optimal policy
- optimization problems
- bayesian networks
- maximum likelihood
- machine learning
- neural network
- database
- bayesian learning
- discrete optimization
- optimization strategies
- efficient optimization
- information technology
- bayesian inference
- constrained optimization
- bayesian estimation
- markov decision process