PIEFS: Physics-Informed Eigenfunction Features with Learnable Scaling

DOI10.48550/arXiv.2607.03692MaRDI QIDQ7283952arXivFDO

Authors Alexander Tarakanov, Varvara Nazarenko, Timur Lidzhiev

Publication date 4 July 2026

Published in arXiv

Tldr PIEFS is a neural framework that learns adaptive coordinates under empirical Gram orthogonality with learnable scaling, yielding task-adaptive Dirichlet-regularized coordinates rather than eigenfunctions of a fixed supervision-independent operator.

Full work available at URL https://doi.org/10.48550/arXiv.2607.03692

arXiv classification

cs.LGmath.NAmath.ST

This page was built for publication: PIEFS: Physics-Informed Eigenfunction Features with Learnable Scaling

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7283952)