{"id":"arxiv/2610.33784","url":"https://arxiv.org/abs/2610.33784","tags":["cs.LG","math.NA"],"score":0,"title":"Greedy Orthogonal Polynomial Manifolds with Lipschitz Control for Dimensionality Reduction and Reduced Modeling","domain":"cs.LG","source":"arxiv","authors":["Tomoki Koike","Benjamin Peherstorfer","Elizabeth Qian"],"reviews":0,"subject":"cs.LG","summary":"Polynomial manifolds provide nonlinear low-dimensional representations through polynomial decoders. We propose to represent the polynomial terms of the decoder using features that are orthonormal on the training data. This empirical orthogonality enables a greedy algorithm, Greedy-OPM, which jointly selects encoder modes and polynomial features while minimizing the reconstruction error under a prescribed bound on the decoder's Lipschitz constant. Numerical experiments show that this Lipschitz constraint improves generalization away from the training data while retaining high reconstruction…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-10-24","reactions":{}}