Opportunities and Challenges in Deep Learning Adversarial Robustness: A Survey

Abstract

As we seek to deploy machine learning models beyond virtual and controlled domains, it is critical to analyze not only the accuracy or the fact that it works most of the time, but if such a model is truly robust and reliable. This paper studies strategies to implement adversary robustly trained algorithms towards guaranteeing safety in machine learning algorithms. We provide a taxonomy to classify adversarial attacks and defenses, formulate the Robust Optimization problem in a min-max setting and divide it into 3 subcategories, namely: Adversarial (re)Training, Regularization Approach, and Certified Defenses. We survey the most recent and important results in adversarial example generation, defense mechanisms with adversarial (re)Training as their main defense against perturbations. We also survey mothods that add regularization terms that change the behavior of the gradient, making it harder for attackers to achieve their objective. Alternatively, we've surveyed methods which formally derive certificates of robustness by exactly solving the optimization problem or by approximations using upper or lower bounds. In addition, we discuss the challenges faced by most of the recent algorithms presenting future research perspectives.

20 pages, 9 figures, submited to IEEE Transactions on Knowledge and Data Engineering

References in corpus (36)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• A Survey on Deep Learning in Medical Image Analysis
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• End to End Learning for Self-Driving Cars
• Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
• Conditional Image Synthesis With Auxiliary Classifier GANs
• ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
• Provable defenses against adversarial examples via the convex outer adversarial polytope
• Towards Deep Neural Network Architectures Robust to Adversarial Examples
• Using Deep Learning and Google Street View to Estimate the Demographic Makeup of the US
• Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey
• Fast is better than free: Revisiting adversarial training
• Adversarial Attacks and Defences: A Survey
• Evaluating Robustness of Neural Networks with Mixed Integer Programming
• Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients
• Robust Large Margin Deep Neural Networks
• Matching matched filtering with deep networks in gravitational-wave astronomy
• Exploring the Landscape of Spatial Robustness
• DARTS: Deceiving Autonomous Cars with Toxic Signs
• Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approach
• UPSET and ANGRI : Breaking High Performance Image Classifiers
• Deep Defense: Training DNNs with Improved Adversarial Robustness
• MACER: Attack-free and Scalable Robust Training via Maximizing Certified Radius
• Unravelling Robustness of Deep Learning based Face Recognition Against Adversarial Attacks
• Metric Learning for Adversarial Robustness
• Robust Decision Trees Against Adversarial Examples
• Human Action Performance using Deep Neuro-Fuzzy Recurrent Attention Model
• Triple Wins: Boosting Accuracy, Robustness and Efficiency Together by Enabling Input-Adaptive Inference
• Query-Efficient Black-box Adversarial Examples (superceded)
• Fooling a Real Car with Adversarial Traffic Signs
• Breaking certified defenses: Semantic adversarial examples with spoofed robustness certificates
• Improved Network Robustness with Adversary Critic
• EMPIR: Ensembles of Mixed Precision Deep Networks for Increased Robustness against Adversarial Attacks
• A Multi-Layer K-means Approach for Multi-Sensor Data Pattern Recognition in Multi-Target Localization
• Statistically Robust Neural Network Classification
• An Adaptive View of Adversarial Robustness from Test-time Smoothing Defense

Cited by in corpus (10)

• Recent Advances in Adversarial Training for Adversarial Robustness
• Learning from Few Samples: A Survey
• Adversarial Machine Learning in Wireless Communications using RF Data: A Review
• Techniques for Symbol Grounding with SATNet
• Asymptotic Behavior of Adversarial Training in Binary Classification
• Adversarial Robustness in Deep Learning: Attacks on Fragile Neurons
• MixDefense: A Defense-in-Depth Framework for Adversarial Example Detection Based on Statistical and Semantic Analysis
• Effect of backdoor attacks over the complexity of the latent space distribution
• Adaptive Clustering of Robust Semantic Representations for Adversarial Image Purification
• Investigating the Robustness of Artificial Intelligent Algorithms with Mixture Experiments