Loss Functions for Top-k Error: Analysis and Insights

TL;DR

This paper analyzes and compares top-k error optimization methods in large-scale classification, introduces new top-k loss functions, and demonstrates their effectiveness and efficiency across multiple datasets.

Contribution

The paper provides a comprehensive evaluation of existing top-k methods and introduces novel loss functions with efficient optimization schemes for improved performance.

Findings

Softmax loss performs competitively for all top-k errors.

New top-k losses improve specific top-k performance.

Proposed methods are faster to train than softmax.

Abstract

In order to push the performance on realistic computer vision tasks, the number of classes in modern benchmark datasets has significantly increased in recent years. This increase in the number of classes comes along with increased ambiguity between the class labels, raising the question if top-1 error is the right performance measure. In this paper, we provide an extensive comparison and evaluation of established multiclass methods comparing their top-k performance both from a practical as well as from a theoretical perspective. Moreover, we introduce novel top-k loss functions as modifications of the softmax and the multiclass SVM losses and provide efficient optimization schemes for them. In the experiments, we compare on various datasets all of the proposed and established methods for top-k error optimization. An interesting insight of this paper is that the softmax loss yields competitive top-k performance for all k simultaneously. For a specific top-k error, our new top-k losses lead typically to further improvements while being faster to train than the softmax.