Chapter

Emergent Discovery
This podcast discusses how discovering new things can be an emergent property of some other objective that quantifies capability, and how multi-agent interaction can lead to understanding that other beings in the world have their own goals, intentions, and thoughts.
Clips
The principles that work for our daily lives are not sufficient to understand medicine and biology.
1:13:04 - 1:16:29 (03:24)
Summary
The principles that work for our daily lives are not sufficient to understand medicine and biology. Training AI systems to interact with humans and to learn from data is a promising but challenging approach.
ChapterEmergent Discovery
Episode#108 – Sergey Levine: Robotics and Machine Learning
PodcastLex Fridman Podcast
This podcast explores whether discovering new things about the world can actually be an emergent property of some other quantifiable objective.
1:16:29 - 1:21:18 (04:49)
Summary
This podcast explores whether discovering new things about the world can actually be an emergent property of some other quantifiable objective. It asks whether there is a role for human curiosity in itself being the reward for discovering new things.
ChapterEmergent Discovery
Episode#108 – Sergey Levine: Robotics and Machine Learning
PodcastLex Fridman Podcast
The difficulty of finding the right reward function in reinforcement learning and how defining an objective to avoid "weird stuff" can mitigate this challenge.
1:21:18 - 1:23:17 (01:58)
Summary
The difficulty of finding the right reward function in reinforcement learning and how defining an objective to avoid "weird stuff" can mitigate this challenge.