{"id":"arxiv/2607.00669","url":"https://arxiv.org/abs/2607.00669","tags":["math.NA","stat.ML"],"score":0,"title":"Convolutional Symmetric AutoEncoders: enhancing latent stability via differential geometry","domain":"math.NA","source":"arxiv","authors":["G. Li Causi","N. Tonicello","L. Magri","G. Rozza"],"reviews":0,"subject":"math.NA","summary":"Autoencoders (AEs) have emerged as powerful tools for non-linear dimensionality reduction, often surpassing traditional linear methods such as Proper Orthogonal Decomposition (POD) in scenarios characterized by slowly decaying Kolmogorov $n$-widths. In the realm of Reduced-Order Modelling (ROM), these models are increasingly utilized to learn low-dimensional representations of solution manifolds associated with parametric Partial Differential Equations (PDEs). However, the high expressivity of AEs presents a challenge: although trained networks typically minimize reconstruction error, they…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-01","reactions":{}}