Multi-Class Classification from Noisy-Similarity-Labeled Data

TL;DR

This paper introduces a novel method for training multi-class classifiers using only noisy similarity labels by modeling noise with a transition matrix, enabling noise-free classification despite label noise.

Contribution

It proposes a new approach that estimates a noise transition matrix from noisy similarity data to learn accurate classifiers without clean labels.

Findings

Outperforms state-of-the-art methods on benchmark noisy datasets

Effectively models noise to improve classification accuracy

Provides theoretical justification for generalization capability

Abstract

A similarity label indicates whether two instances belong to the same class while a class label shows the class of the instance. Without class labels, a multi-class classifier could be learned from similarity-labeled pairwise data by meta classification learning. However, since the similarity label is less informative than the class label, it is more likely to be noisy. Deep neural networks can easily remember noisy data, leading to overfitting in classification. In this paper, we propose a method for learning from only noisy-similarity-labeled data. Specifically, to model the noise, we employ a noise transition matrix to bridge the class-posterior probability between clean and noisy data. We further estimate the transition matrix from only noisy data and build a novel learning system to learn a classifier which can assign noise-free class labels for instances. Moreover, we theoretically justify how our proposed method generalizes for learning classifiers. Experimental results demonstrate the superiority of the proposed method over the state-of-the-art method on benchmark-simulated and real-world noisy-label datasets.