{"id":"arxiv/2606.19895","url":"https://arxiv.org/abs/2606.19895","tags":["math.NA","cs.LG"],"score":0,"title":"A fast direct solver based neural network for solving PDEs","domain":"math.NA","source":"arxiv","authors":["Jashwanth Reddy Kadaru","Vaishnavi Gujjula"],"reviews":0,"subject":"math.NA","summary":"The matrices arising from large scale $N$-body problems can be efficiently represented using hierarchical matrices, whose key idea is that the admissible off-diagonal sub-matrices can be well approximated by low-rank matrices across a hierarchy of matrix partitions. HODLR (Hierarchical Off-Diagonal Low-Rank) matrices are a subclass of hierarchical matrices in which all off-diagonal submatrices at every level of a recursive binary partition are low-rank. In this article, we present a neural network that learns the inverse operation of HODLR matrices based on the fast direct solver for HODLR…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-18","reactions":{}}