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

Most methods for explaining black-box classifiers (e.g. on tabular data,images, or time series) rely on measuring the impact that removing/perturbingfeatures has on the model output. This forces the explanation language to matchthe classifier's feature space. However, when dealing with graph data, in whichthe basic features correspond to the edges describing the graph structure, thismatching between features space and explanation language might not beappropriate. Decoupling the feature space (edges) from a desired high-levelexplanation language (such as motifs) is thus a major challenge towardsdeveloping actionable explanations for graph classification tasks. In thispaper we introduce GRAPHSHAP, a Shapley-based approach able to providemotif-based explanations for identity-aware graph classifiers, assuming noknowledge whatsoever about the model or its training data: the only requirementis that the classifier can be queried as a black-box at will. For the sake ofcomputational efficiency we explore a progressive approximation strategy andshow how a simple kernel can efficiently approximate explanation scores, thusallowing GRAPHSHAP to scale on scenarios with a large explanation space (i.e.large number of motifs). We showcase GRAPHSHAP on a real-world brain-networkdataset consisting of patients affected by Autism Spectrum Disorder and acontrol group. Our experiments highlight how the classification provided by ablack-box model can be effectively explained by few connectomics patterns.

GRAPHSHAP: Explaining Identity-Aware Graph Classifiers Through the Language of Motifs