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) have emerged as highly successful tools forgraph-related tasks. However, real-world problems involve very large graphs,and the compute resources needed to fit GNNs to those problems grow rapidly.Moreover, the noisy nature and size of real-world graphs cause GNNs to over-fitif not regularized properly. Surprisingly, recent works show that large graphsoften involve many redundant components that can be removed withoutcompromising the performance too much. This includes node or edge removalsduring inference through GNNs layers or as a pre-processing step thatsparsifies the input graph. This intriguing phenomenon enables the developmentof state-of-the-art GNNs that are both efficient and accurate. In this paper,we take a further step towards demystifying this phenomenon and propose asystematic method called Locality-Sensitive Pruning (LSP) for graph pruningbased on Locality-Sensitive Hashing. We aim to sparsify a graph so that similarlocal environments of the original graph result in similar environments in theresulting sparsified graph, which is an essential feature for graph-relatedtasks. To justify the application of pruning based on local graph properties,we exemplify the advantage of applying pruning based on locality propertiesover other pruning strategies in various scenarios. Extensive experiments onsynthetic and real-world datasets demonstrate the superiority of LSP, whichremoves a significant amount of edges from large graphs without compromisingthe performance, accompanied by a considerable acceleration.

LSP : Acceleration and Regularization of Graph Neural Networks via Locality Sensitive Pruning of Graphs