The Vapnik-Chervonenkis (VC) dimension is a measure of the capacity of a set of functions that can be learned by a binary classification algorithm. It is defined as the maximum number of points that the algorithm can perfectly classify, and was introduced by Vladimir Vapnik and Alexey Chervonenkis. The capacity of a model is related to how complex it can be, with higher capacity models being more wiggly and thus able to fit training data better but making more errors on other points.
UC Berkeley
Fall 2013
This course investigates the mathematical principles behind data and information analysis. It brings together concepts from statistics, optimization, and computer science, with a focus on large deviation inequalities, and convex analysis. It's tailored towards advanced graduate students who wish to incorporate these theories into their research.
No concepts data
+ 20 more concepts