Top-k Multiclass SVM

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.

NIPS 2015

Cited by in corpus (12)

• Tensor Normalization and Full Distribution Training
• Set-valued classification -- overview via a unified framework
• Learning by Minimizing the Sum of Ranked Range
• The Limited Multi-Label Projection Layer
• sigmoidF1: A Smooth F1 Score Surrogate Loss for Multilabel Classification
• Learning with Fenchel-Young Losses
• The Z-loss: a shift and scale invariant classification loss belonging to the Spherical Family
• T${}_k$ML-AP: Adversarial Attacks to Top-$k$ Multi-Label Learning
• On the Consistency of Top-k Surrogate Losses
• Classification of sparse binary vectors
• Projections onto the canonical simplex with additional linear inequalities
• Sum of Ranked Range Loss for Supervised Learning