English

Implicitly Maximizing Margins with the Hinge Loss

Machine Learning 2020-06-26 v1 Machine Learning

Abstract

A new loss function is proposed for neural networks on classification tasks which extends the hinge loss by assigning gradients to its critical points. We will show that for a linear classifier on linearly separable data with fixed step size, the margin of this modified hinge loss converges to the 2\ell_2 max-margin at the rate of O(1/t)\mathcal{O}( 1/t ). This rate is fast when compared with the O(1/logt)\mathcal{O}(1/\log t) rate of exponential losses such as the logistic loss. Furthermore, empirical results suggest that this increased convergence speed carries over to ReLU networks.

Keywords

Cite

@article{arxiv.2006.14286,
  title  = {Implicitly Maximizing Margins with the Hinge Loss},
  author = {Justin Lizama},
  journal= {arXiv preprint arXiv:2006.14286},
  year   = {2020}
}
R2 v1 2026-06-23T16:37:06.747Z