Chapter
The Importance of Having an Admissible Set of Functions in Neural Networks
31:47 - 37:08 (05:20)
By having an admissible set of functions, we can create a subset of functions that we can use to create architecture. Closed form solutions use this set of functions, which is not the same as piecewise linear functions. Neural networks use this set of functions, making it better to have a set of functions than to rely on piecewise linear functions.
Clips
The discovery of good predicates is the essence of human-level intelligence that AI seeks.
31:47 - 32:54 (01:06)
Summary
The discovery of good predicates is the essence of human-level intelligence that AI seeks. While deep learning is a popular method, it still relies on mediocre predicates, leaving much room for improvement.
ChapterThe Importance of Having an Admissible Set of Functions in Neural Networks
EpisodeVladimir Vapnik: Predicates, Invariants, and the Essence of Intelligence
PodcastLex Fridman Podcast
The use of deep learning is to construct architectures that can converge towards a function which generalizes efficiently rather than relying on piecewise linear functions which are limited in their scope.
32:54 - 35:28 (02:34)
Summary
The use of deep learning is to construct architectures that can converge towards a function which generalizes efficiently rather than relying on piecewise linear functions which are limited in their scope.
ChapterThe Importance of Having an Admissible Set of Functions in Neural Networks
EpisodeVladimir Vapnik: Predicates, Invariants, and the Essence of Intelligence
PodcastLex Fridman Podcast
Neural networks are just a small subset of a larger space of functions, and when creating a network architecture, you want to use an admissible set of functions that are useful for your specific problem.
35:28 - 37:08 (01:39)
Summary
Neural networks are just a small subset of a larger space of functions, and when creating a network architecture, you want to use an admissible set of functions that are useful for your specific problem. It is important to understand this larger space of functions in order to create better networks.