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 Convolutional Networks (GCNs) have emerged as the state-of-the-artmethod for graph-based learning tasks. However, training GCNs at scale is stillchallenging, hindering both the exploration of more sophisticated GCNarchitectures and their applications to real-world large graphs. While it mightbe natural to consider graph partition and distributed training for tacklingthis challenge, this direction has only been slightly scratched the surface inthe previous works due to the limitations of existing designs. In this work, wefirst analyze why distributed GCN training is ineffective and identify theunderlying cause to be the excessive number of boundary nodes of eachpartitioned subgraph, which easily explodes the memory and communication costsfor GCN training. Furthermore, we propose a simple yet effective method dubbedBNS-GCN that adopts random Boundary-Node-Sampling to enable efficient andscalable distributed GCN training. Experiments and ablation studiesconsistently validate the effectiveness of BNS-GCN, e.g., boosting thethroughput by up to 16.2x and reducing the memory usage by up to 58%, whilemaintaining a full-graph accuracy. Furthermore, both theoretical and empiricalanalysis show that BNS-GCN enjoys a better convergence than existingsampling-based methods. We believe that our BNS-GCN has opened up a newparadigm for enabling GCN training at scale. The code is available athttps://github.com/RICE-EIC/BNS-GCN.

BNS-GCN: Efficient Full-Graph Training of Graph Convolutional Networks with Partition-Parallelism and Random Boundary Node Sampling