Historical Foundations of RLHF
Nathan explores the deep-rooted connections between reinforcement learning and ancient philosophy, highlighting how the von Neumann Morgenstern utility theorem shapes our understanding of preferences. He raises critical questions about the influence of internet-sourced data versus curated datasets on AI models, emphasizing the need for insights from social sciences to better comprehend human preferences in modern AI.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
802: In Case You Missed It in June 2024 — with Jon Krohn (@JonKrohnLearns)
Related Questions