The VC dimension of constraint-based grammars

VC dimension of ellipsoids

We will establish that the vc dimension of the class of d-dimensional ellipsoids is (d +3d)/2, and that maximum likelihood estimate with N -component d-dimensional Gaussian mixture models induces a geometric class having vc dimension at least N(d + 3d)/2.

Teaching Dimension versus VC Dimension

In this report, we give a brief survey of various results relating the Teaching Dimension and VC-Dimension. The concept of Teaching Dimension was first introduced by Goldman and Kearns, 1995 and Sinohara and Miyano, 1991. In this model, an algorithm tries to learn the hidden concept c from examples, called the teaching set, which uniquely identifies c from the rest of the concepts in the concep...

On the Eeective Vc Dimension

The very idea of an \EEective Vapnik Chervonenkis (VC) dimension" (Vapnik, Levin and Le Cun, 1993) relies on the hypothesis that the relation between the generalization error and the number of training examples can be expressed by a formula algebraically similar to the VC bound. This hypothesis calls for a serious discussion since the traditional VC bound widely overestimates the generalization...

Inapproximability of VC Dimension and Littlestone's Dimension

We study the complexity of computing the VC Dimension and Littlestone’s Dimension. Given an explicit description of a finite universe and a concept class (a binary matrix whose (x,C)-th entry is 1 iff element x belongs to concept C), both can be computed exactly in quasipolynomial time (n ). Assuming the randomized Exponential Time Hypothesis (ETH), we prove nearly matching lower bounds on the ...

Feature-based Grammars as Constraint Grammars