Reinforcement Learning Insights
Jeremie explains the concept of reinforcement learning from human feedback (RLHF) as a method to refine powerful text autocomplete systems. He discusses the training process, which begins with extensive pre-training on vast amounts of text, followed by human evaluators providing feedback to improve the system's outputs. The conversation delves into the ongoing debate within the AI safety community about whether these adjustments genuinely address the underlying risks or merely serve as temporary fixes.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
668: GPT-4: Apocalyptic stepping stone? — with Jeremie Harris
Related Questions