AXRP - the AI X-risk Research Podcast cover image

33 - RLHF Problems with Scott Emmons

AXRP - the AI X-risk Research Podcast

NOTE

Optimal Policy vs Arbitrary Teaching Policy

The focus is on the optimal policy compared to an arbitrary teaching policy, where the goal is to achieve the task efficiently. The discussion revolves around the idea that the deception of a policy can be influenced by the ability of humans to form true beliefs, raising questions about the choice between basing policies on human beliefs or striving for optimal outcomes.

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner