Chii chinonzi Kusimbisa Kudzidza kubva kuHuman Feedback (RLHF)? Tsanangudzo | AI Understanding

Tsanangudzo

Nzira yekudzidzisa inoshandisa zviratidzo zvinofarirwa nevanhu kugadzira maitiro emuenzaniso.

Mashoko anoenderana

Kusimbisa Kudzidza

Kudzidziswa nemasaini masaini apo mumiririri anodzidza zviito zvinowedzera kudzoka kwenguva refu.

Modhi inoburitsa zvibodzwa zvichibva pazviratidzo zvekuda, zvinowanzo shandiswa mumapaipi eRLHF.

DPO (Direct Preference Optimization)

Nzira yekudzidzisa iyo inonatsa-tuni modhi yakananga pamapeya ekuda pasina kuda mubairo wakasiyana.

Kuenderera mberi nekudzidza

Nzira dzekudzidzisa dzinoita kuti modhi irambe ichidzidza kubva kune itsva data pasina kukanganwa ruzivo rwekare.

Vashoma-Kupfura Kudzidza

Kudzidza kana kugadzirisa maitiro kubva kunhamba shoma yemienzaniso.

Kudzikira kwekuita nekufamba kwenguva sezvo mamiriro epasirese chaiwo anosiyana kubva pakufungidzira kudzidziswa.

Dzidza zvakawanda mumadhairekitori edu emahara

Deep Learning Reinforcement Learning Machine Learning Basics Supervised Learning

Onawo

Retrieval Red Teaming Recommendation System Robustness Recall Safety Filter RAG (Retrieval-Augmented Generation)Scaling Law

Bhurawuza yakazara AI Glossary Ongorora ese emahara AI madhairekitori