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

Existing knowledge distillation methods on graph neural networks (GNNs) arealmost offline, where the student model extracts knowledge from a powerfulteacher model to improve its performance. However, a pre-trained teacher modelis not always accessible due to training cost, privacy, etc. In this paper, wepropose a novel online knowledge distillation framework to resolve thisproblem. Specifically, each student GNN model learns the extracted localstructure from another simultaneously trained counterpart in an alternatingtraining procedure. We further develop a cross-layer distillation strategy byaligning ahead one student layer with the layer in different depth of anotherstudent model, which theoretically makes the structure information spread overall layers. Experimental results on five datasets including PPI,Coauthor-CS/Physics and Amazon-Computer/Photo demonstrate that the studentperformance is consistently boosted in our collaborative training frameworkwithout the supervision of a pre-trained teacher model. In addition, we alsofind that our alignahead technique can accelerate the model convergence speedand its effectiveness can be generally improved by increasing the studentnumbers in training. Code is available:https://github.com/GuoJY-eatsTG/Alignahead

Alignahead: Online Cross-Layer Knowledge Extraction on Graph Neural Networks