ERASE: Error-Resilient Representation Learning on Graphs for Label Noise Tolerance

TL;DR

ERASE is a novel method that enhances the robustness of graph neural networks against label noise by learning error-tolerant representations through a combination of label correction, propagation, and coding rate reduction.

Contribution

The paper introduces ERASE, a new approach that improves graph representation learning under noisy labels by decoupling label propagation and incorporating denoising and error resilience techniques.

Findings

ERASE outperforms baseline methods across various noise levels.

The method demonstrates strong scalability and robustness.

Experimental results confirm improved node classification accuracy.

Abstract

Deep learning has achieved remarkable success in graph-related tasks, yet this accomplishment heavily relies on large-scale high-quality annotated datasets. However, acquiring such datasets can be cost-prohibitive, leading to the practical use of labels obtained from economically efficient sources such as web searches and user tags. Unfortunately, these labels often come with noise, compromising the generalization performance of deep networks. To tackle this challenge and enhance the robustness of deep learning models against label noise in graph-based tasks, we propose a method called ERASE (Error-Resilient representation learning on graphs for lAbel noiSe tolerancE). The core idea of ERASE is to learn representations with error tolerance by maximizing coding rate reduction. Particularly, we introduce a decoupled label propagation method for learning representations. Before training, noisy labels are pre-corrected through structural denoising. During training, ERASE combines prototype pseudo-labels with propagated denoised labels and updates representations with error resilience, which significantly improves the generalization performance in node classification. The proposed method allows us to more effectively withstand errors caused by mislabeled nodes, thereby strengthening the robustness of deep networks in handling noisy graph data. Extensive experimental results show that our method can outperform multiple baselines with clear margins in broad noise levels and enjoy great scalability. Codes are released at https://github.com/eraseai/erase.