Improving the Robustness of Deep Neural Networks via Stability Training

In this paper we address the issue of output instability of deep neural networks: small perturbations in the visual input can significantly distort the feature embeddings and output of a neural network. Such instability affects many deep architectures with state-of-the-art performance on a wide rang...

Full description

Saved in:
Bibliographic Details
Published in2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) pp. 4480 - 4488
Main Authors Zheng, Stephan, Yang Song, Leung, Thomas, Goodfellow, Ian
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2016
Subjects
Data models
Feature extraction
Neural networks
Robustness
Stability analysis
Training
Visualization
Online AccessGet full text

Cover

More Information
Summary:In this paper we address the issue of output instability of deep neural networks: small perturbations in the visual input can significantly distort the feature embeddings and output of a neural network. Such instability affects many deep architectures with state-of-the-art performance on a wide range of computer vision tasks. We present a general stability training method to stabilize deep networks against small input distortions that result from various types of common image processing, such as compression, rescaling, and cropping. We validate our method by stabilizing the state of-the-art Inception architecture [11] against these types of distortions. In addition, we demonstrate that our stabilized model gives robust state-of-the-art performance on largescale near-duplicate detection, similar-image ranking, and classification on noisy datasets.
ISSN:1063-6919
DOI:10.1109/CVPR.2016.485