Top-k Multiclass SVM

TL;DR

This paper introduces a top-k multiclass SVM that directly optimizes for top-k accuracy, improving performance in image classification tasks with many classes by using a convex upper bound and efficient optimization.

Contribution

It presents a novel top-k multiclass SVM formulation with a tight convex upper bound and a fast projection-based optimization scheme, enhancing top-k accuracy.

Findings

Consistent improvements in top-k accuracy across five datasets.

Efficient optimization via projection onto the top-k simplex.

Effective handling of class ambiguity in large-scale image classification.

Abstract

Class ambiguity is typical in image classification problems with a large number of classes. When classes are difficult to discriminate, it makes sense to allow k guesses and evaluate classifiers based on the top-k error instead of the standard zero-one loss. We propose top-k multiclass SVM as a direct method to optimize for top-k performance. Our generalization of the well-known multiclass SVM is based on a tight convex upper bound of the top-k error. We propose a fast optimization scheme based on an efficient projection onto the top-k simplex, which is of its own interest. Experiments on five datasets show consistent improvements in top-k accuracy compared to various baselines.