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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Graph neural networks (GNNs) have been successfully exploited in graphanalysis tasks in many real-world applications. The competition between attackand defense methods also enhances the robustness of GNNs. In this competition,the development of adversarial training methods put forward higher requirementfor the diversity of attack examples. By contrast, most attack methods withspecific attack strategies are difficult to satisfy such a requirement. Toaddress this problem, we propose GraphAttacker, a novel generic graph attackframework that can flexibly adjust the structures and the attack strategiesaccording to the graph analysis tasks. GraphAttacker generates adversarialexamples through alternate training on three key components: the multi-strategyattack generator (MAG), the similarity discriminator (SD), and the attackdiscriminator (AD), based on the generative adversarial network (GAN).Furthermore, we introduce a novel similarity modification rate SMR to conduct astealthier attack considering the change of node similarity distribution.Experiments on various benchmark datasets demonstrate that GraphAttacker canachieve state-of-the-art attack performance on graph analysis tasks of nodeclassification, graph classification, and link prediction, no matter theadversarial training is conducted or not. Moreover, we also analyze the uniquecharacteristics of each task and their specific response in the unified attackframework. The project code is available athttps://github.com/honoluluuuu/GraphAttacker.

GraphAttacker: A General Multi-Task GraphAttack Framework