Id: arxiv/2606.27982 Url: https://arxiv.org/abs/2606.27982 Tags: math.DS Score: 0 Title: PRONE: Petrov-Galerkin Operator Learning Unifies DMD, SINDy & Koopmanism Domain: math.DS Source: arxiv Authors: Matthew J. Colbrook, April Herwig, J. Nathan Kutz Reviews: 0 Subject: math.DS Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-06-26 Page: https://delphi.science/r/arxiv/2606.27982 Data-driven dynamics often asks how to linearize a nonlinear system. We ask instead: which observables should be advanced, and where should their futures live? This leads to Petrov Regression Of Nonlinear Evolution (PRONE), a Petrov--Galerkin regression framework based on $ Ψ(\mathbf{X})K \approx Φ(\mathbf{Y}), $ with distinct trial and test dictionaries. In this form, DMD, EDMD, SINDy, Koopman regression, sparse regression, and low-rank regression become variants of one construction: different dictionaries, weights, and constraints. We keep the linear algebra of Koopman learning, but drop…