Search

Your search keyword '"Luo, Dan"' showing total 8,076 results
Start Over Author "Luo, Dan"
8,076 results on '"Luo, Dan"'

123. Memory Defense: More Robust Classification via a Memory-Masking Autoencoder

Author

Adhikarla, Eashan, Luo, Dan, and Davison, Brian D.

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning
Abstract

Many deep neural networks are susceptible to minute perturbations of images that have been carefully crafted to cause misclassification. Ideally, a robust classifier would be immune to small variations in input images, and a number of defensive approaches have been created as a result. One method would be to discern a latent representation which could ignore small changes to the input. However, typical autoencoders easily mingle inter-class latent representations when there are strong similarities between classes, making it harder for a decoder to accurately project the image back to the original high-dimensional space. We propose a novel framework, Memory Defense, an augmented classifier with a memory-masking autoencoder to counter this challenge. By masking other classes, the autoencoder learns class-specific independent latent representations. We test the model's robustness against four widely used attacks. Experiments on the Fashion-MNIST & CIFAR-10 datasets demonstrate the superiority of our model. We make available our source code at GitHub repository: https://github.com/eashanadhikarla/MemDefense, Comment: 11 pages

Published
2022

Catalog

Books, media, physical & digital resources
See catalog results