Searcharxiv⌕ Search

arXiv subjects

Felipe Tome

Publications and source records attributed to Felipe Tome.

1 recordsLinked to original sources

Learning to Select Sparse Linear Solvers with Convolutional Neural Networks

We investigate convolutional-neural-network (CNN) based automatic solver selection for sparse linear systems, a problem of practical importance where solver performance varies dramatically with matrix sparsity structure. A CNN is trained to regress log timing ratios between candidate solvers using spy plot images of matrix nonzero structure as input, achieving strong prediction accuracy on both held-out SuiteSparse matrices and real-world matrices from high-performance simulation codes.

cs.LG↗