Abstract:Network data in real-world tends to be error-prone due to incomplete sampling, imperfect measurements, etc.; this in turn results in inaccurate results when performing network analysis or modeling, such as node classification and link prediction, on these flawed networks. In this paper, we aim to reconstruct a reliable network from a flawed, undirected, unweighted network, a process referred to network enhancement. More specifically, network enhancement aims to detect the noisy links that are observed in the network but should not exist in the real world, as well as to predict the missing links that do indeed exist in the real world yet remain unobserved. While some attempts have been made to detect either noisy links or missing links, few of these works have considered unifying these two tasks, even though they are inter-dependent and capable of mutually boosting each others’ performance. In this paper, we therefore propose E-Net, an end-to-end graph neural network model, to leverage the mutual influence of these two tasks in order to achieve both goals more effectively. On one hand, detecting noisy links can benefit the performance of missing link prediction, while on the other hand, predicting missing links can provide indirect supervision for detecting noisy link detection when the labels of these noisy links are unavailable. Moreover, by proposing a subgraph extraction mechanism based on random walk with restart, the model can be scaled up to large networks and is able to preserve the local and global structural characteristics. The experimental results on several types of large networks demonstrate that the proposed model obtains an improvement of 10.7 percent on average in terms of F1 for predicting missing links, along with an average of 3.7 percent improvement in terms of precision for detecting noisy links compared with the state-of-the-art baselines.

USER: Unsupervised Structural Entropy-Based Robust Graph Neural Network

Unsupervised Adversarially-Robust Representation Learning on Graphs

Robust Network Enhancement from Flawed Networks.

How Robust Are Graph Neural Networks to Structural Noise?

GraphRARE: Reinforcement Learning Enhanced Graph Neural Network with Relative Entropy

Robust graph neural networks with Dirichlet regularization and residual connection

You Can't Ignore Either: Unifying Structure and Feature Denoising for Robust Graph Learning

Self-supervised robust Graph Neural Networks against noisy graphs and noisy labels

Reliable Representations Make A Stronger Defender: Unsupervised Structure Refinement for Robust GNN

RSGNN: residual structure graph neural network

Graph Structure Estimation Neural Networks

GARNET: Reduced-Rank Topology Learning for Robust and Scalable Graph Neural Networks

GNN-Ensemble: Towards Random Decision Graph Neural Networks

Graph Stochastic Neural Networks for Semi-supervised Learning.

Robustness of Graph Classification: failure modes, causes, and noise-resistant loss in Graph Neural Networks

LSEGNN: Encode Local Topology Structure in Graph Neural Networks

Status-Aware Signed Heterogeneous Network Embedding With Graph Neural Networks

Robust Node Representation Learning via Graph Variational Diffusion Networks

A Unified Framework of Graph Information Bottleneck for Robustness and Membership Privacy

Robust Graph Neural Networks via Unbiased Aggregation

Graph Structure Reshaping Against Adversarial Attacks on Graph Neural Networks