Observe Locally, Classify Globally: Using GNNs to Identify Sparse Matrix Structure

TL;DR

This paper presents a graph neural network-based framework for classifying sparse matrix structures efficiently using local samples, achieving high accuracy to optimize matrix format selection for computations.

Contribution

The authors introduce a novel GNN-based framework that classifies sparse matrix structures from local samples, addressing the challenge of structure identification without full data access.

Findings

Achieves 97% classification accuracy on representative sparse matrix shapes.

Effectively infers global matrix structure from local features.

Framework can be extended to other matrix structures with user-defined generators.

Abstract

The performance of sparse matrix computation highly depends on the matching of the matrix format with the underlying structure of the data being computed on. Different sparse matrix formats are suitable for different structures of data. Therefore, the first challenge is identifying the matrix structure before the computation to match it with an appropriate data format. The second challenge is to avoid reading the entire dataset before classifying it. This can be done by identifying the matrix structure through samples and their features. Yet, it is possible that global features cannot be determined from a sampling set and must instead be inferred from local features. To address these challenges, we develop a framework that generates sparse matrix structure classifiers using graph convolutional networks. The framework can also be extended to other matrix structures using user-provided generators. The approach achieves 97% classification accuracy on a set of representative sparse matrix shapes.