Understanding Deep Learning
Yann explores the parallels between thermodynamics and intelligence, questioning the theoretical foundations of deep learning. He highlights the surprising success of larger neural networks despite traditional statistical theories suggesting otherwise. The conversation reveals the importance of intuitive understanding in machine learning, cautioning against overly complex models that may lead to ineffective engineering.In this clip
From this podcast
The Gradient
Yann LeCun on his Start in Research and Self-Supervised Learning
Related Questions
Why are neural networks hard to explain as discussed in the episode Vladimir Vapnik: Statistical Learning | Lex Fridman Podcast #5 and the clip Deep Learning Critique?
Why are neural networks hard to explain as discussed in the episode Vladimir Vapnik: Statistical Learning | Lex Fridman Podcast #5 and the clip Deep Learning Critique?
Is deep learning a scientific practice as discussed in the episode Yann LeCun on his Start in Research and Self-Supervised Learning and the clip Deep Learning's Alchemical Nature?