2607.06048

Total: 1

#1 Separation Capacity of Scattering Networks on Low-Dimensional Datasets [PDF] [Copy] [Kimi] [REL]

Authors: Konstantin Häberle, Helmut Bölcskei

We aim to identify scattering network architectures that maximize the separation capacity on data with low intrinsic dimension. The networks we consider employ a fixed monomial nonlinearity and no pooling, so that the only design variable is the frame generated by the network filters. For data modeled as rectifiable sets, we first characterize and bound the separation capacity of general feature extractors in terms of the geometry of the dataset. We then particularize to scattering networks and obtain two design criteria: (i) the filters should meet the data on sufficiently many frequencies, and (ii) the matrices coupling the frame to the geometry of the data should be well-conditioned.

Subjects: Machine Learning , Information Theory , Machine Learning , Classical Analysis and ODEs

Publish: 2026-07-07 09:25:58 UTC