3

Improving Reward Learning by Estimating Annotator Expertise
Reinforcement learning from human feedback (RLHF) has been used successfully to teach robots tasks that are difficult to specify …
Pavel Czempin, Rachel Freedman, Ellen Novoseller, Vernon Lawhern, Cameron Allen, Erdem Bıyık
PDF Cite
In-Context Generalization to New Tasks From Unlabeled Observation Data
Large pretrained models in natural language processing and computer vision have achieved impressive capabilities by training on vast …
Anthony Liang, Pavel Czempin, Yutai Zhou, Stephen Tu, Erdem Bıyık
PDF
In-Context Generalization to New Tasks From Unlabeled Observation Data