Adapt, Agree, Aggregate: Semi-Supervised Ensemble Labeling for Graph Convolutional Networks

2025-03-22

AI Summary

To address the vulnerability of Graph Convolutional Networks (GCNs) to error accumulation and confirmation bias under noisy graph structures in semi-supervised node classification, this paper proposes a multi-view ensemble-based robust training framework. Methodologically, it constructs multiple augmented graph views, ensembles several GCN models, and introduces an adaptive consensus mechanism that dynamically selects high-confidence samples via confidence-thresholding to collaboratively generate robust pseudo-labels. The core contribution is the first introduction of an “ensemble-consensus-driven” adaptive pseudo-labeling paradigm, which explicitly mitigates confirmation bias and error propagation. Extensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art semi-supervised approaches across multiple real-world graph benchmarks, particularly maintaining high accuracy and strong robustness under sparse labeling and edge-noise perturbations.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

In this paper, we propose a novel framework that combines ensemble learning with augmented graph structures to improve the performance and robustness of semi-supervised node classification in graphs. By creating multiple augmented views of the same graph, our approach harnesses the"wisdom of a diverse crowd", mitigating the challenges posed by noisy graph structures. Leveraging ensemble learning allows us to simultaneously achieve three key goals: adaptive confidence threshold selection based on model agreement, dynamic determination of the number of high-confidence samples for training, and robust extraction of pseudo-labels to mitigate confirmation bias. Our approach uniquely integrates adaptive ensemble consensus to flexibly guide pseudo-label extraction and sample selection, reducing the risks of error accumulation and improving robustness. Furthermore, the use of ensemble-driven consensus for pseudo-labeling captures subtle patterns that individual models often overlook, enabling the model to generalize better. Experiments on several real-world datasets demonstrate the effectiveness of our proposed method.

Problem

• Improving semi-supervised node classification in graphs
• Mitigating noisy graph structure challenges via ensemble learning
• Enhancing robustness through adaptive pseudo-label extraction

Innovation

• Ensemble learning with augmented graph structures
• Adaptive confidence threshold selection
• Ensemble-driven consensus for pseudo-labeling