• search
    search
  • reviewers
    reviewers
  • feeds
    feeds
  • assignments
    assignments
  • settings
  • logout

Gibbs Sampling from Human Feedback: A Provable KL- constrained Framework for RLHF.

Wei XiongHanze DongChenlu YeHan ZhongNan JiangTong Zhang
Published in: CoRR (2023)
Keyphrases
  • gibbs sampling
  • computer vision
  • parameter estimation
  • image reconstruction
  • active learning
  • least squares
  • topic models