{"id":"arxiv/2609.17137","url":"https://arxiv.org/abs/2609.17137","tags":["math.NA"],"score":0,"title":"Latent Inversion of Material Coefficients from Boundary Data via Finite Tests and Neural Surrogates","domain":"math.NA","source":"arxiv","authors":["Erik Burman","Mats G. Larson","Karl Larsson","Jonatan Vallin"],"reviews":0,"subject":"math.NA","summary":"We reconstruct a spatially varying material coefficient in a scalar elliptic equation from finitely many boundary excitations, each producing a full Dirichlet trace. To mitigate the ill-posedness and the cost of repeated PDE solves, we restrict the coefficient to a low-dimensional family, specified analytically or learned from samples, and solve the inverse problem in its latent coordinates. We consider a \\(C^1\\) parametrization with \\(m\\) latent coordinates and full-rank derivative at a reference point. If the continuous linearized Neumann-to-Dirichlet map is injective on the corresponding…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-15","reactions":{}}