Reinforcement Learning from Human Feedback (RLHF)

A Simplified Explanation

Maybe you’ve heard about this technique, but you haven’t completely understood it, especially the PPO part. This explanation might help.

We will focus on text-to-text language models 📝, such as GPT-3, BLOOM, and T5. Models like BERT, which are encoder-only, are not addressed.

This blog post is an adaptation of the gist from the same author.

Reinforcement Learning from Human Feedback (RLHF) has been successfully applied in ChatGPT, hence its major increase in popularity. 📈

RLHF …

artificial intelligence bert blog bloom encoder feedback focus gpt gpt-3 human human feedback language language models large language models machine learning part ppo reinforcement reinforcement learning rlhf simplified text text-to-text transformers