AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
2min chapter
Accelerating Intelligence with AI-Generating Algorithms with Jeff Clune - #602
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Is There an Outer Loop Meta Reward Function?
I think instead we probably need to be looking for that outer loop meta reward. That basically allows broadly RIT a system that can learn forever and not learn to memorize white noise patterns but instead learns things that we find useful. Now that's a really simplistic it won't work in practice but that is I think where a lot of the research should be focused which is can we figure out those kind of general reward functions plug it into an AIGA and good things continue to happen forever. Could we create a computer algorithm that was worth running for a billion years that continues to innovate and delight surprise and be creative? If we can then we've made a lot of progress towards some really fascinating
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