MLLGFeb 10, 2021

An Exact Solver for the Weston-Watkins SVM Subproblem

arXiv:2102.05640v21 citations
AI Analysis

This work addresses a computational bottleneck in multiclass SVM training, offering incremental improvements for machine learning practitioners.

The paper tackles the subproblem in Weston-Watkins SVM solvers by developing an exact algorithm using a novel reparametrization, resulting in significant speed-ups for large numbers of classes and enabling a proof of linear convergence.

Recent empirical evidence suggests that the Weston-Watkins support vector machine is among the best performing multiclass extensions of the binary SVM. Current state-of-the-art solvers repeatedly solve a particular subproblem approximately using an iterative strategy. In this work, we propose an algorithm that solves the subproblem exactly using a novel reparametrization of the Weston-Watkins dual problem. For linear WW-SVMs, our solver shows significant speed-up over the state-of-the-art solver when the number of classes is large. Our exact subproblem solver also allows us to prove linear convergence of the overall solver.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes