DeepSearch: A Simple and Effective Blackbox Attack for Deep Neural Networks
arXiv:1910.06296
Abstract
Although deep neural networks have been very successful in image-classification tasks, they are prone to adversarial attacks. To generate adversarial inputs, there has emerged a wide variety of techniques, such as black- and whitebox attacks for neural networks. In this paper, we present DeepSearch, a novel fuzzing-based, query-efficient, blackbox attack for image classifiers. Despite its simplicity, DeepSearch is shown to be more effective in finding adversarial inputs than state-of-the-art blackbox approaches. DeepSearch is additionally able to generate the most subtle adversarial inputs in comparison to these approaches.
References in corpus (30)
- Wide Residual Networks
- ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
- Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
- DeepXplore: Automated Whitebox Testing of Deep Learning Systems
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning
- Provable defenses against adversarial examples via the convex outer adversarial polytope
- On Evaluating Adversarial Robustness
- Certified Defenses against Adversarial Examples
- Guiding Deep Learning System Testing using Surprise Adequacy
- Adversarial Risk and the Dangers of Evaluating Against Weak Attacks
- Evaluating Robustness of Neural Networks with Mixed Integer Programming
- DLFuzz: Differential Fuzzing Testing of Deep Learning Systems
- Formal Security Analysis of Neural Networks using Symbolic Intervals
- An approach to reachability analysis for feed-forward ReLU neural networks
- Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priors
- Defensive Distillation is Not Robust to Adversarial Examples
- Testing Deep Neural Networks
- A Dual Approach to Scalable Verification of Deep Networks
- Simple Black-box Adversarial Attacks
- MagNet and "Efficient Defenses Against Adversarial Attacks" are Not Robust to Adversarial Examples
- A Unified View of Piecewise Linear Neural Network Verification
- On the Robustness of the CVPR 2018 White-Box Adversarial Example Defenses
- Evaluating and Understanding the Robustness of Adversarial Logit Pairing
- Reachable Set Computation and Safety Verification for Neural Networks with ReLU Activations
- Combinatorial Testing for Deep Learning Systems
- Parsimonious Black-Box Adversarial Attacks via Efficient Combinatorial Optimization
- Symbolic Execution for Deep Neural Networks
- Bypassing Feature Squeezing by Increasing Adversary Strength
- On the Limitation of MagNet Defense against -based Adversarial Examples
- The Efficacy of SHIELD under Different Threat Models