Transversal pooling networks improve stability under input transformations
Transversal Pooling Neural Networks
Machine Learning
Summary
Many computer learning tasks need to recognize objects even when they are slightly moved or transformed, but still detect bigger changes. The authors introduce transversal pooling neural networks, which extend a common image-processing step called max pooling to more complex transformations like rotations and stretches. They prove these networks behave predictably when inputs are changed slightly and test them on small datasets and predicting hurricane strength changes. This new approach helps computer systems better manage images or data that are shifted or transformed.
What this means in practice
- •For weather forecasting teams: Improve predictions of tropical cyclone intensification by using stable neural network models that handle small changes in input data.
- •For machine learning engineers: Build neural networks that retain sensitivity to major data changes while being robust to minor transformations in low-data environments.
Authors
Emily J. King, Dustin G. Mixon, Michael Perlmutter, Lander 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.