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) ability to generalize across complexdistributions is crucial for real-world applications. However, prior researchhas primarily focused on specific types of distribution shifts, such as largergraph size, or inferred shifts from constructed data environments, which ishighly limited when confronted with multiple and nuanced distribution shifts.For instance, in a social graph, a user node might experience increasedinteractions and content alterations, while other user nodes encounter distinctshifts. Neglecting such complexities significantly impedes generalization. Toaddress it, we present GraphMETRO, a novel framework that enhances GNNgeneralization under complex distribution shifts in both node and graph-leveltasks. Our approach employs a mixture-of-experts (MoE) architecture with agating model and expert models aligned in a shared representation space. Thegating model identifies key mixture components governing distribution shifts,while each expert generates invariant representations w.r.t. a mixturecomponent. Finally, GraphMETRO aggregates representations from multiple expertsto generate the final invariant representation. Our experiments on syntheticand realworld datasets demonstrate GraphMETRO's superiority andinterpretability. To highlight, GraphMETRO achieves state-of-the-artperformances on four real-world datasets from GOOD benchmark, outperforming thebest baselines on WebKB and Twitch datasets by 67% and 4.2%, respectively.

GraphMETRO: Mitigating Complex Distribution Shifts in GNNs via Mixture of Aligned Experts