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

Temporal Graph Learning, which aims to model the time-evolving nature ofgraphs, has gained increasing attention and achieved remarkable performancerecently. However, in reality, graph structures are often incomplete and noisy,which hinders temporal graph networks (TGNs) from learning informativerepresentations. Graph contrastive learning uses data augmentation to generateplausible variations of existing data and learn robust representations.However, rule-based augmentation approaches may be suboptimal as they lacklearnability and fail to leverage rich information from downstream tasks. Toaddress these issues, we propose a Time-aware Graph Structure Learning (TGSL)approach via sequence prediction on temporal graphs, which learns better graphstructures for downstream tasks through adding potential temporal edges. Inparticular, it predicts time-aware context embedding based on previouslyobserved interactions and uses the Gumble-Top-K to select the closest candidateedges to this context embedding. Additionally, several candidate samplingstrategies are proposed to ensure both efficiency and diversity. Furthermore,we jointly learn the graph structure and TGNs in an end-to-end manner andperform inference on the refined graph. Extensive experiments on temporal linkprediction benchmarks demonstrate that TGSL yields significant gains for thepopular TGNs such as TGAT and GraphMixer, and it outperforms other contrastivelearning methods on temporal graphs. We release the code athttps://github.com/ViktorAxelsen/TGSL.

Time-aware Graph Structure Learning via Sequence Prediction on Temporal Graphs