A few days ago, the data science community engaged in an intense debate when AI legend and Chief AI Scientist at Meta, Yann LeCun made some remarks about the fact that ChatGPT was not particularly innovative. Although controversial in light of the almost magical capabilities of ChatGPT, the statement is rooted in the fact that many of the ideas behind ChatGPT have been around for a while, and ChatGPT has been more the result of clever implementation than breakthrough research. One of the key enablers of the ChatGPT magic can be traced back to 2017 under the obscure name of reinforcement learning with human feedback(RLHF).

Large language models(LLMs) have become one of the most interesting environments for applying modern reinforcement learning(RL) techniques. While LLMs are great at deriving knowledge from vast amounts of text, RL can help to translate that knowledge into actions. That has been the secret behind RLHF...