RLHF

say it: “AR-EL-AYCH-EF” /ˌɑrˌɛlˌeɪtʃˈɛf/

Training models on human ratings of their answers.

What is RLHF?

RLHF (reinforcement learning from human feedback) trains models to give answers people prefer, using ratings of responses.

Level: Master · More AI Basics terms at this level

Updated