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 formidable resources forprocessing graph-based information across diverse applications. While theexpressive power of GNNs has traditionally been examined in the context ofgraph-level tasks, their potential for node-level tasks, such as nodeclassification, where the goal is to interpolate missing node labels from theobserved ones, remains relatively unexplored. In this study, we investigate theproficiency of GNNs for such classifications, which can also be cast as afunction interpolation problem. Explicitly, we focus on ascertaining theoptimal configuration of weights and layers required for a GNN to successfullyinterpolate a band-limited function over Euclidean cubes. Our findingshighlight a pronounced efficiency in utilizing GNNs to generalize a bandlimitedfunction within an $\varepsilon$-error margin. Remarkably, achieving this tasknecessitates only $O_d((\log\varepsilon^{-1})^d)$ weights and$O_d((\log\varepsilon^{-1})^d)$ training samples. We explore how this criterionstacks up against the explicit constructions of currently available NeuralNetworks (NNs) designed for similar tasks. Significantly, our result isobtained by drawing an innovative connection between the GNN structures andclassical sampling theorems. In essence, our pioneering work marks a meaningfulcontribution to the research domain, advancing our understanding of thepractical GNN applications.

Superiority of GNN over NN in generalizing bandlimited functions