CVCRLGMLDec 26, 2017

The Robust Manifold Defense: Adversarial Training using Generative Models

arXiv:1712.09196v5176 citationsHas Code
Originality Incremental advance
AI Analysis

This addresses security vulnerabilities in machine learning systems, particularly for image classification, with incremental improvements in robustness.

The paper tackles the problem of adversarial attacks on image classifiers by proposing a new attack method that exploits generative models to find adversarial examples, reducing Defense-GAN accuracy to 3% and achieving the most robust known MNIST classifier against PGD attacks.

We propose a new type of attack for finding adversarial examples for image classifiers. Our method exploits spanners, i.e. deep neural networks whose input space is low-dimensional and whose output range approximates the set of images of interest. Spanners may be generators of GANs or decoders of VAEs. The key idea in our attack is to search over latent code pairs to find ones that generate nearby images with different classifier outputs. We argue that our attack is stronger than searching over perturbations of real images. Moreover, we show that our stronger attack can be used to reduce the accuracy of Defense-GAN to 3\%, resolving an open problem from the well-known paper by Athalye et al. We combine our attack with normal adversarial training to obtain the most robust known MNIST classifier, significantly improving the state of the art against PGD attacks. Our formulation involves solving a min-max problem, where the min player sets the parameters of the classifier and the max player is running our attack, and is thus searching for adversarial examples in the {\em low-dimensional} input space of the spanner. All code and models are available at \url{https://github.com/ajiljalal/manifold-defense.git}

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes