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 paradigm of vertical federated learning (VFL), where institutionscollaboratively train machine learning models via combining each other's localfeature or label information, has achieved great success in applications tofinancial risk management (FRM). The surging developments of graphrepresentation learning (GRL) have opened up new opportunities for FRMapplications under FL via efficiently utilizing the graph-structured datagenerated from underlying transaction networks. Meanwhile, transactioninformation is often considered highly sensitive. To prevent data leakageduring training, it is critical to develop FL protocols with formal privacyguarantees. In this paper, we present an end-to-end GRL framework in the VFLsetting called VESPER, which is built upon a general privatization schemetermed perturbed message passing (PMP) that allows the privatization of manypopular graph neural architectures.Based on PMP, we discuss the strengths andweaknesses of specific design choices of concrete graph neural architecturesand provide solutions and improvements for both dense and sparse graphs.Extensive empirical evaluations over both public datasets and an industrydataset demonstrate that VESPER is capable of training high-performance GNNmodels over both sparse and dense graphs under reasonable privacy budgets.

Privacy-preserving design of graph neural networks with applications to vertical federated learning