Keywords: deep neural networks, visual cortex, invariance, data augmentation
TL;DR: We propose data augmentation invariance: a simple, yet effective and efficient way of learning robust features by adding a layer-wise invariance objective in the loss function.
Abstract: Deep convolutional neural networks trained for image object categorization have shown remarkable similarities with representations found across the primate ventral visual stream. Yet, artificial and biological networks still exhibit important differences. Here we investigate one such property: increasing invariance to identity-preserving image transformations found along the ventral stream. Despite theoretical evidence that invariance should emerge naturally from the optimization process, we present empirical evidence that the activations of convolutional neural networks trained for object categorization are not robust to identity-preserving image transformations commonly used in data augmentation. As a solution, we propose data augmentation invariance, an unsupervised learning objective which improves the robustness of the learned representations by promoting the similarity between the activations of augmented image samples. Our results show that this approach is a simple, yet effective and efficient (10 % increase in training time) way of increasing the invariance of the models while obtaining similar categorization performance.
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:1906.04547/code)
Original Pdf: pdf
7 Replies
Loading