Authors
Emily J. KingDustin G. MixonMichael PerlmutterLander Ver Hoef
Abstract
Many learning tasks require stability to small transformations while retaining sensitivity to larger ones. We introduce \emph{transversal pooling neural networks}, which generalize spatial max pooling to affine group actions. We establish equivariance to a chosen subgroup and derive explicit stability bounds for individual pooled wavelet coefficients under affine perturbations of the input. Experimentally, we demonstrate the utility of our networks in low-data environments and for predicting tropical cyclone intensification.
Paper Information
- arXiv ID:
- 2609.36237v1
- Published:
Related Categories
cs.LGmath.FA