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 (GNN) show great promise in problems dealing withgraph-structured data. One of the unique points of GNNs is their flexibility toadapt to multiple problems, which not only leads to wide applicability, butalso poses important challenges when finding the best model or accelerationtechnique for a particular problem. An example of such challenges resides inthe fact that the accuracy or effectiveness of a GNN model or accelerationtechnique generally depends on the structure of the underlying graph. In thispaper, in an attempt to address the problem of graph-dependent acceleration, wepropose ProGNNosis, a data-driven model that can predict the GNN training timeof a given GNN model running over a graph of arbitrary characteristics byinspecting the input graph metrics. Such prediction is made based on aregression that was previously trained offline using a diverse synthetic graphdataset. In practice, our method allows making informed decisions on whichdesign to use for a specific problem. In the paper, the methodology to buildProGNNosis is defined and applied for a specific use case, where it helps todecide which graph representation is better. Our results show that ProGNNosishelps achieve an average speedup of 1.22X over randomly selecting a graphrepresentation in multiple widely used GNN models such as GCN, GIN, GAT, orGraphSAGE.

ProGNNosis: A Data-driven Model to Predict GNN Computation Time Using Graph Metrics