Chapter
The Science of Admissible Set of Functions
14:42 - 23:05 (08:22)
This podcast episode discusses the concept of admissible set of functions in the field of science and mathematics. The goal is to create a set of functions with a small capacity or VC dimension, leading to a more effective and accurate representation.
Clips
This podcast discusses the significance of choosing relevant predicates in machine learning, and how theoretical and empirical descriptions should coincide for accurate outcomes.
14:42 - 16:04 (01:22)
Summary
This podcast discusses the significance of choosing relevant predicates in machine learning, and how theoretical and empirical descriptions should coincide for accurate outcomes.
ChapterThe Science of Admissible Set of Functions
EpisodeVladimir Vapnik: Statistical Learning
PodcastLex Fridman Podcast
The English proverb "swims like a dog and quacks like a dog" exemplifies the ambiguity present in language, as it requires a certain level of knowledge and understanding to fully appreciate its meaning.
16:05 - 17:03 (00:57)
Summary
The English proverb "swims like a dog and quacks like a dog" exemplifies the ambiguity present in language, as it requires a certain level of knowledge and understanding to fully appreciate its meaning. The underlying knowledge base needed to comprehend this proverb and others like it is accumulated over time through experience and exposure to various cultural contexts.
ChapterThe Science of Admissible Set of Functions
EpisodeVladimir Vapnik: Statistical Learning
PodcastLex Fridman Podcast
This podcast episode explores the concept of predicate variables in machine learning and how they are used to decrease the admissible set of functions.
17:03 - 18:30 (01:27)
Summary
This podcast episode explores the concept of predicate variables in machine learning and how they are used to decrease the admissible set of functions. It is an essential tool in understanding the essence of dog, among other things.
ChapterThe Science of Admissible Set of Functions
EpisodeVladimir Vapnik: Statistical Learning
PodcastLex Fridman Podcast
In machine learning, an admissible set of functions is a set with small capacity or diversity, and a good function can be found within this set.
18:32 - 19:32 (00:59)
Summary
In machine learning, an admissible set of functions is a set with small capacity or diversity, and a good function can be found within this set. VC theory refers to a way of measuring the complexity of an admissible set, which is important when selecting a function for a machine to use in making predictions.
ChapterThe Science of Admissible Set of Functions
EpisodeVladimir Vapnik: Statistical Learning
PodcastLex Fridman Podcast
The smaller the VC dimension, the less training data is needed for learning.
19:32 - 21:20 (01:47)
Summary
The smaller the VC dimension, the less training data is needed for learning. The goal of learning is to create an admissible set of functions with a small VC dimension, which contain good functions that can be picked up with a small amount of observations.
ChapterThe Science of Admissible Set of Functions
EpisodeVladimir Vapnik: Statistical Learning
PodcastLex Fridman Podcast
The process of creating an admissible set of functions that is invariant involves looking at properties of the training data, identifying general and special types of predicates and ensuring their values coincide with the expected values of the model.
21:20 - 23:05 (01:44)
Summary
The process of creating an admissible set of functions that is invariant involves looking at properties of the training data, identifying general and special types of predicates and ensuring their values coincide with the expected values of the model.