ConceptAI & ML
VC dimension
A measure of the capacity of a statistical classification model, used in generalization bounds.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Regularization3Data snooping2Generalization bound2Complexity of a set of objects1Effective VC dimension1Fat margin1Feature selection vs VC dimension1Generalization in deep learning1Generalization result for support vector machines1Lilstone dimension1Rule of Thumb for VC dimension examples1Testable learning1
Lecturers
6Each in their own terms, from the easiest level.
Yaser Abu-Mostafa
The VC dimension of a hypothesis set is defined as the most points that the hypothesis set can shatter.Phillip Isola
The cardinality of the set of dichotomies that a model class can shatter over a given number of training points.Ilya Sutskever
A measure of the capacity of a statistical classification model, invented to handle parameters with infinite precision.Santosh Vempala
A measure of the capacity of a statistical classification algorithm, determining the sample complexity needed to learn a distribution.Nina Balcan
The Vapnik-Chervonenkis dimension measures the capacity of a hypothesis space of verifiers.
Lectures that assume it
9Connections
29The lectures that define it or assume it, and the concepts next to it.
Concepts10
Lectures19
- The VC Dimension
- Support Vector Machines
- Error and Noise
- Bias-Variance Tradeoff
- Neural Networks
- Learning Representations: A Challenge for Learning Theory, COLT 2013
Show 13 moreShow fewer
- An Observation on Generalization and Unsupervised Compression
- Generalization Theory
- Theory of Modern AI: Learning Theoretic, Game Theoretic, and Algorithmic Perspectives
- Red Points and Blue Points
- The Linear Model II
- Overfitting
- Regularization
- Validation
- Radial Basis Functions
- Three Learning Principles
- Efficient Algorithms for Reliable Machine Learning
- Hacker's Guide to Deep Learning
- Computational Thinking on Learning Models