English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Node classification using Graph Neural Networks (GNNs) has been widelyapplied in various real-world scenarios. However, in recent years, compellingevidence emerges that the performance of GNN-based node classification maydeteriorate substantially by topological perturbation, such as randomconnections or adversarial attacks. Various solutions, such as topologicaldenoising methods and mechanism design methods, have been proposed to developrobust GNN-based node classifiers but none of these works can fully address theproblems related to topological perturbations. Recently, the Bayesian labeltransition model is proposed to tackle this issue but its slow convergence maylead to inferior performance. In this work, we propose a new label inferencemodel, namely LInDT, which integrates both Bayesian label transition andtopology-based label propagation for improving the robustness of GNNs againsttopological perturbations. LInDT is superior to existing label transitionmethods as it improves the label prediction of uncertain nodes by utilizingneighborhood-based label propagation leading to better convergence of labelinference. Besides, LIndT adopts asymmetric Dirichlet distribution as a prior,which also helps it to improve label inference. Extensive experiments on fivegraph datasets demonstrate the superiority of LInDT for GNN-based nodeclassification under three scenarios of topological perturbations.

Robust Node Classification on Graphs: Jointly from Bayesian Label Transition and Topology-based Label Propagation