A self-supervised multigrid graph network with triplet sampling from the Fill-Path Theorem and an end-max chain loss reduces fill-ins and speeds up LU factorization on SuiteSparse matrices.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Self-Supervised Learning for Sparse Matrix Reordering
A self-supervised multigrid graph network with triplet sampling from the Fill-Path Theorem and an end-max chain loss reduces fill-ins and speeds up LU factorization on SuiteSparse matrices.