{"id":"arxiv/2607.17723","url":"https://arxiv.org/abs/2607.17723","tags":["math.NA"],"score":0,"title":"A domain decomposition online-learning-enhanced nonlinear elimination preconditioner","domain":"math.NA","source":"arxiv","authors":["Pai Zhang","Linyan Gu","Li Luo"],"reviews":0,"subject":"math.NA","summary":"Nonlinearly preconditioned inexact Newton methods form an effective class of solvers for large-scale nonlinear algebraic systems arising from the discretization of partial differential equations. A central challenge in nonlinear elimination (NE) preconditioning is the reliable identification of the slowly converging components to be eliminated. Existing selection strategies often rely on problem-specific physical information or user-tuned thresholds applied directly to the raw nonlinear residual, which may contain irregular oscillatory structures near stagnation regions, making the selected…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-20","reactions":{}}