RLHF
2017
Train a reward model from human preferences, then optimise the policy against it. Christiano 2017, InstructGPT 2022. Turned GPT-3 into ChatGPT.
Also: Reinforcement learning from human feedback
Sources: AI4U editorial.
2017
Train a reward model from human preferences, then optimise the policy against it. Christiano 2017, InstructGPT 2022. Turned GPT-3 into ChatGPT.
Also: Reinforcement learning from human feedback
Sources: AI4U editorial.