ZOO: Zeroth Order Optimization Based Black-box Attacks to Deep Neural Networks without Training Substitute Models (2017-08-14T00:00:00.000000Z)

TL;DR

An effective black-box attack that also only has access to the input (images) and the output (confidence scores) of a targeted DNN is proposed, sparing the need for training substitute models and avoiding the loss in attack transferability.

Abstract

Deep neural networks (DNNs) are one of the most prominent technologies of our time, as they achieve state-of-the-art performance in many machine learning tasks, including but not limited to image classification, text mining, and speech processing. However, recent research on DNNs has indicated ever-increasing concern on the robustness to adversarial examples, especially for security-critical tasks such as traffic sign identification for autonomous driving. Studies have unveiled the vulnerability of a well-trained DNN by demonstrating the ability of generating barely noticeable (to both human and machines) adversarial images that lead to misclassification. Furthermore, researchers have shown that these adversarial images are highly transferable by simply training and attacking a substitute model built upon the target model, known as a black-box attack to DNNs. Similar to the setting of training substitute models, in this paper we propose an effective black-box attack that also only has access to the input (images) and the output (confidence scores) of a targeted DNN. However, different from leveraging attack transferability from substitute models, we propose zeroth order optimization (ZOO) based attacks to directly estimate the gradients of the targeted DNN for generating adversarial examples. We use zeroth order stochastic coordinate descent along with dimension reduction, hierarchical attack and importance sampling techniques to efficiently attack black-box models. By exploiting zeroth order optimization, improved attacks to the targeted DNN can be accomplished, sparing the need for training substitute models and avoiding the loss in attack transferability. Experimental results on MNIST, CIFAR10 and ImageNet show that the proposed ZOO attack is as effective as the state-of-the-art white-box attack (e.g., Carlini and Wagner's attack) and significantly outperforms existing black-box attacks via substitute models.

References44 items

Robust Physical-World Attacks on Deep Learning Models

Robust Physical-World Attacks on Machine Learning Models

Efficient Defenses Against Adversarial Attacks

Adversarial Examples, Uncertainty, and Transfer Testing Robustness in Gaussian Process Hybrid Deep Networks

Towards Deep Learning Models Resistant to Adversarial Attacks

Feature Squeezing Mitigates and Detects Carlini/Wagner Adversarial Examples

Analysis of universal adversarial perturbations

Black-Box Attacks against RNN based Malware Detection Algorithms

Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods

Ensemble Adversarial Training: Attacks and Defenses

Extending Defensive Distillation

Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks

Detecting Adversarial Samples from Artifacts

On the (Statistical) Detection of Adversarial Examples

On Detecting Adversarial Perturbations

Delving into Transferable Adversarial Examples and Black-box Attacks

Adversarial Machine Learning at Scale

Universal Adversarial Perturbations

Safety Verification of Deep Neural Networks

Towards Evaluating the Robustness of Neural Networks

Adversarial examples in the physical world

Adversarial Perturbations Against Deep Neural Networks for Malware Classification

A Comprehensive Linear Speedup Analysis for Asynchronous Stochastic Parallel Optimization from Zeroth-Order to First-Order

Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples

Crafting adversarial input sequences for recurrent neural networks

Improving the Robustness of Deep Neural Networks via Stability Training

Practical Black-Box Attacks against Machine Learning

Rethinking the Inception Architecture for Computer Vision

Random Gradient-Free Minimization of Convex Functions

The Limitations of Deep Learning in Adversarial Settings

Robust Convolutional Neural Networks under Adversarial Noise

DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks

Distillation as a Defense to Adversarial Perturbations Against Deep Neural Networks