A solvable high-dimensional model where nonlinear autoencoders learn structure invisible to PCA while test loss misaligns with generalization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Mendes, Vicente Conde, Bardone, Lorenzo, Koller, Cédric, Moreira, Jorge Medina, Erba, Vittorio, Troiani, Emanuele, Zdeborová, Lenka
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Disordered Systems and Neural Networks
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2602.10680