Instance-dependent Label-noise Learning under a Structural Causal Model

TL;DR

This paper introduces a causal-model-based generative approach to handle instance-dependent label noise in deep learning, improving classifier performance by modeling data structure and noise transition matrices.

Contribution

It proposes a novel method leveraging structural causal models to better identify label noise transition matrices, enhancing robustness against label errors.

Findings

Outperforms state-of-the-art methods on synthetic datasets

Effective in real-world label-noise scenarios

Improves classifier accuracy by modeling causal structures

Abstract

Label noise will degenerate the performance of deep learning algorithms because deep neural networks easily overfit label errors. Let X and Y denote the instance and clean label, respectively. When Y is a cause of X, according to which many datasets have been constructed, e.g., SVHN and CIFAR, the distributions of P(X) and P(Y|X) are entangled. This means that the unsupervised instances are helpful to learn the classifier and thus reduce the side effect of label noise. However, it remains elusive on how to exploit the causal information to handle the label noise problem. In this paper, by leveraging a structural causal model, we propose a novel generative approach for instance-dependent label-noise learning. In particular, we show that properly modeling the instances will contribute to the identifiability of the label noise transition matrix and thus lead to a better classifier. Empirically, our method outperforms all state-of-the-art methods on both synthetic and real-world label-noise datasets.