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) are recently getting much attention inbioinformatics and chemoinformatics as a state-of-the-art machine learningapproach with high accuracy. GCNs process convolutional operations along withgraph structures, and GPUs are used to process enormous operations includingsparse-dense matrix multiplication (SpMM) when the graph structure is expressedas an adjacency matrix with sparse matrix format. However, the SpMM operationon small graph, where the number of nodes is tens or hundreds, hardly exploitshigh parallelism or compute power of GPU. Therefore, SpMM becomes a bottleneckof training and inference in GCNs applications. In order to improve theperformance of GCNs applications, we propose new SpMM algorithm especially forsmall sparse matrix and Batched SpMM, which exploits high parallelism of GPU byprocessing multiple SpMM operations with single CUDA kernel. To the best of ourknowledge, this is the first work of batched approach for SpMM. We evaluatedthe performance of the GCNs application on TSUBAME3.0 implementing NVIDIA TeslaP100 GPU, and our batched approach shows significant speedups of up to 1.59xand 1.37x in training and inference, respectively.

Batched Sparse Matrix Multiplication for Accelerating Graph Convolutional Networks