AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
3min chapter
Episode 29: Jim Fan, NVIDIA, on foundation models for embodied agents, scaling data, and why prompt engineering will become irrelevant
Generally Intelligent
Is There a Refinement Learning From Human Feedback?
I think the reinforcement learning from human feedback is very exciting, not just because it delivered chat GPT, but also because it's like a general purpose paradigm to align powerful AI systems with human tech. The reason problem generic will not be relevant for ever is because of our trust. We have to kind of coerce the model to give us what we want by typing out a very natural sentences and to essentially trick the model into solving the task. But if we can just tell it all to what task we want to solve in the most natural language that we can come up with then this whole problem engineering business is no longer needed. Even my grandma can use charge Bt and solve her task on
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode