Chapter
The Real Power of Neural Networks
15:41 - 21:59 (06:18)
Understanding the power of neural networks does not come from their inscrutability but from their differentiability.
Clips
The speaker argues that many failures in automation could have been prevented if those involved were more suspicious and tested technology more rigorously.
15:41 - 16:26 (00:44)
Summary
The speaker argues that many failures in automation could have been prevented if those involved were more suspicious and tested technology more rigorously.
ChapterThe Real Power of Neural Networks
Episode#155 – Max Tegmark: AI and Physics
PodcastLex Fridman Podcast
This podcast episode discusses the issue of understanding neural networks, reinforcing the point that these systems are black boxes similar to humans' brains, and sharing several accidents caused by robots due to a lack of understanding of human differences.
16:26 - 19:14 (02:48)
Summary
This podcast episode discusses the issue of understanding neural networks, reinforcing the point that these systems are black boxes similar to humans' brains, and sharing several accidents caused by robots due to a lack of understanding of human differences.
ChapterThe Real Power of Neural Networks
Episode#155 – Max Tegmark: AI and Physics
PodcastLex Fridman Podcast
The real power of neural networks does not come from their inscrutability or mystery, but rather from their differentiability.
19:14 - 21:59 (02:45)
Summary
The real power of neural networks does not come from their inscrutability or mystery, but rather from their differentiability. In this podcast, the hosts discuss the idea that the power of neural networks stems from their ability to differentiate and the lack of mysticism surrounding them.