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

How can we augment a dynamic graph for improving the performance of dynamicgraph neural networks? Graph augmentation has been widely utilized to boost thelearning performance of GNN-based models. However, most existing approachesonly enhance spatial structure within an input static graph by transforming thegraph, and do not consider dynamics caused by time such as temporal locality,i.e., recent edges are more influential than earlier ones, which remainschallenging for dynamic graph augmentation. In this work, we propose TiaRa(Time-aware Random Walk Diffusion), a novel diffusion-based method foraugmenting a dynamic graph represented as a discrete-time sequence of graphsnapshots. For this purpose, we first design a time-aware random walk proximityso that a surfer can walk along the time dimension as well as edges, resultingin spatially and temporally localized scores. We then derive our diffusionmatrices based on the time-aware random walk, and show they become enhancedadjacency matrices that both spatial and temporal localities are augmented.Throughout extensive experiments, we demonstrate that TiaRa effectivelyaugments a given dynamic graph, and leads to significant improvements indynamic GNN models for various graph datasets and tasks.

Time-aware Random Walk Diffusion to Improve Dynamic Graph Learning