Observe Locally, Classify Globally: Using GNNs to Identify Sparse Matrix Structure
This addresses the challenge of optimizing sparse matrix computation performance for practitioners in high-performance computing, though it is incremental as it builds on existing GNN methods.
The paper tackles the problem of matching sparse matrix formats to underlying data structures by developing a framework using graph convolutional networks to classify matrix structures from local features, achieving 97% accuracy on representative shapes.
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.