Chapter
The Search for Small Circuits and Programs in Neural Networks
40:17 - 50:16 (09:59)
The discussion revolves around the idea of neural networks as the search for small circuits and general intelligence as the search for small programs. It explores whether the architecture needed to enable neural networks to reason will look similar to what we have today or require a different approach.
Clips
The speaker discusses the possibility of implementing the mechanism of learning of the brain in neural networks to improve their performance.
40:17 - 43:08 (02:51)
Summary
The speaker discusses the possibility of implementing the mechanism of learning of the brain in neural networks to improve their performance. Back propagation in the brain should be explored to learn something new in how the brain learns.
ChapterThe Search for Small Circuits and Programs in Neural Networks
Episode#94 – Ilya Sutskever: Deep Learning
PodcastLex Fridman Podcast
The search for small circuits, and maybe general intelligence as the search for small programs.
43:08 - 47:37 (04:29)
Summary
The search for small circuits, and maybe general intelligence as the search for small programs. Neural networks are capable of reasoning, but it is uncertain whether the architecture for neural networks to reason will look similar to today's architectures.
ChapterThe Search for Small Circuits and Programs in Neural Networks
Episode#94 – Ilya Sutskever: Deep Learning
PodcastLex Fridman Podcast
This episode discusses how the neural networks' ability to transmit entropy from the data set to parameters can aid them in finding small programs, as well as why there are no successful methodologies for people to find programs manually.
47:37 - 50:16 (02:38)
Summary
This episode discusses how the neural networks' ability to transmit entropy from the data set to parameters can aid them in finding small programs, as well as why there are no successful methodologies for people to find programs manually.