Published on Tue Jun 11 2019

Learning robust visual representations using data augmentation invariance

Alex Hernández-García, Peter König, Tim C. Kietzmann

Deep convolutional neural networks trained for image object categorization have shown remarkable similarities with representations found across the primate ventral visual stream. We propose data augmentation invariance, an unsupervised learning objective which improves the robustness of the learned representations.

0
0
0
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.

Wed May 30 2018
Computer Vision
Why do deep convolutional networks generalize so poorly to small image transformations?
Convolutional Neural Networks (CNNs) are commonly assumed to be invariant to small image transformations. Small translations or rescalings of the input image can drastically change the network's prediction. We show that neither the convolutional architecture nor data augmentation are sufficient to achieve the
0
0
0
Tue Feb 09 2021
Machine Learning
More Is More -- Narrowing the Generalization Gap by Adding Classification Heads
Overfit is a fundamental problem in machine learning in general, and in deep learning in particular. We introduce an architecture enhancement for existing neural network models based on input transformations. Our model can be employed during training time only and then pruned for prediction, resulting in an equivalent architecture.
0
0
0
Wed Dec 18 2013
Neural Networks
Unsupervised feature learning by augmenting single images
When deep learning is applied to visual object recognition, data augmentation is often used to generate additional training data without extra labeling cost. It helps to reduce overfitting and increase the performance of the algorithm.
0
0
0
Fri Jul 27 2018
Computer Vision
Diverse feature visualizations reveal invariances in early layers of deep neural networks
Visualizing features in deep neural networks (DNNs) can help understanding their computations. Many previous studies aimed to visualize the selectivity of individual units. Here we propose a method to discover invariances in the responses of hidden layer units of deep neural networkworks.
0
0
0
Tue Jun 30 2020
Machine Learning
Is Robustness To Transformations Driven by Invariant Neural Representations?
Deep Convolutional Neural Networks (DCNNs) have demonstrated impressive robustness to recognize objects under transformations. A hypothesis to explain such robustness is that DCNNs develop invariant neural representations that remain unaltered when the image is transformed.
0
0
0
Wed Jun 26 2019
Machine Learning
Further advantages of data augmentation on convolutional neural networks
Data augmentation is a popular technique largely used to enhance the training of convolutional neural networks. Many of its benefits are well known by deep learning researchers and practitioners. But its implicit regularization effects remain largely unstudied.
0
0
0