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

In Graph Neural Networks (GNNs), the graph structure is incorporated into thelearning of node representations. This complex structure makes explaining GNNs'predictions become much more challenging. In this paper, we proposePGM-Explainer, a Probabilistic Graphical Model (PGM) model-agnostic explainerfor GNNs. Given a prediction to be explained, PGM-Explainer identifies crucialgraph components and generates an explanation in form of a PGM approximatingthat prediction. Different from existing explainers for GNNs where theexplanations are drawn from a set of linear functions of explained features,PGM-Explainer is able to demonstrate the dependencies of explained features inform of conditional probabilities. Our theoretical analysis shows that the PGMgenerated by PGM-Explainer includes the Markov-blanket of the targetprediction, i.e. including all its statistical information. We also show thatthe explanation returned by PGM-Explainer contains the same set of independencestatements in the perfect map. Our experiments on both synthetic and real-worlddatasets show that PGM-Explainer achieves better performance than existingexplainers in many benchmark tasks.

PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks