Sign rank versus VC dimension

[edit]

Noga Alon, Shay Moran, Amir Yehudayoff ;
29th Annual Conference on Learning Theory, PMLR 49:47-80, 2016.

Abstract

This work studies the maximum possible sign rank of N \times N sign matrices with a given VC dimension d. For d=1, this maximum is three. For d=2, this maximum is \tildeΘ(N^1/2). For d >2, similar but slightly less accurate statements hold. Our lower bounds improve over previous ones by Ben-David et al. and can be interpreted as exhibiting a weakness of kernel-based classifiers. Our upper bounds, on the other hand, can be interpreted as exhibiting the universality of kernel-based classifiers. The lower bounds are obtained by probabilistic constructions, using a theorem of Warren in real algebraic topology. The upper bounds are obtained using a result of Welzl about spanning trees with low stabbing number, and using the moment curve. The upper bound technique is also used to: (i) provide estimates on the number of classes of a given VC dimension, and the number of maximum classes of a given VC dimension – answering a question of Frankl from ’89, and (ii) design an efficient algorithm that provides an O(N/\log(N)) multiplicative approximation for the sign rank (computing the sign rank is equivalent to the existential theory of the reals). We also observe a general connection between sign rank and spectral gaps which is based on Forster’s argument. Consider the N \times N adjacency matrix of a ∆regular graph with a second eigenvalue of absolute value λand ∆≤N/2. We show that the sign rank of the signed version of this matrix is at least ∆/λ. We use this connection to prove the existence of a maximum class C⊆{\pm 1}^N with VC dimension 2 and sign rank \tildeΘ(N^1/2). This answers a question of Ben-David et al. regarding the sign rank of large VC classes. We also describe limitations of this approach, in the spirit of the Alon-Boppana theorem. We further describe connections to communication complexity, geometry, learning theory, and combinatorics.

Related Material