alt Hacker NewsThis kind of approach would generally still need human guidance, otherwise these models might get stuck in weird niche corners of the problem space that would not be relevant to any real world project.
This kind of approach would generally still need human guidance, otherwise these models might get stuck in weird niche corners of the problem space that would not be relevant to any real world project.
We could call this "reinforcement learning from human feedback" (RLHF) :)
https://en.wikipedia.org/wiki/Reinforcement_learning_from_hu...