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

The deployment of Graph Neural Networks (GNNs) within Machine Learning as aService (MLaaS) has opened up new attack surfaces and an escalation in securityconcerns regarding model-centric attacks. These attacks can directly manipulatethe GNN model parameters during serving, causing incorrect predictions andposing substantial threats to essential GNN applications. Traditional integrityverification methods falter in this context due to the limitations imposed byMLaaS and the distinct characteristics of GNN models.  In this research, we introduce a groundbreaking approach to protect GNNmodels in MLaaS from model-centric attacks. Our approach includes acomprehensive verification schema for GNN's integrity, taking into account bothtransductive and inductive GNNs, and accommodating varying pre-deploymentknowledge of the models. We propose a query-based verification technique,fortified with innovative node fingerprint generation algorithms. To deal withadvanced attackers who know our mechanisms in advance, we introduce randomizedfingerprint nodes within our design. The experimental evaluation demonstratesthat our method can detect five representative adversarial model-centricattacks, displaying 2 to 4 times greater efficiency compared to baselines.

Securing Graph Neural Networks in MLaaS: A Comprehensive Realization of Query-based Integrity Verification