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) are popular machine learning models for graphswith many applications across scientific domains. However, GNNs are consideredblack box models, and it is challenging to understand how the model makespredictions. Game theory-based Shapley value approaches are popular explanationmethods in other domains but are not well-studied for graphs. Some studies haveproposed Shapley value-based GNN explanations, yet they have severallimitations: they consider limited samples to approximate Shapley values; somemainly focus on small and large coalition sizes, and they are an order ofmagnitude slower than other explanation methods, making them inapplicable toeven moderate-size graphs. In this work, we propose GNNShap, which providesexplanations for edges since they provide more natural explanations for graphsand more fine-grained explanations. We overcome the limitations by samplingfrom all coalition sizes, parallelizing the sampling on GPUs, and speeding upmodel predictions by batching. GNNShap gives better fidelity scores and fasterexplanations than baselines on real-world datasets.

GNNShap: Fast and Accurate GNN Explanations using Shapley Values