The Alexander-Hirschowitz theorem for neurovarieties
2511.19703v1
math.AG, cs.AI, cs.LG, math.AC
2025-11-26
Авторы:
A. Massarenti, M. Mella
Abstract
We study neurovarieties for polynomial neural networks and fully characterize when they attain the expected dimension in the single-output case. As consequences, we establish non-defectiveness and global identifiability for multi-output architectures.