MaxDumbledore/PapersAboutAdversarialExamples

我关于神经网络对抗样本方面论文的笔记

PapersAboutAdversarialExamples

本仓库仅关注视觉神经网络的对抗样本。

2019年起的大部分论文已经被列出。

目前更新到ICLR 2020 Conference | OpenReview

综述类

2017年以后

• Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey
• Adversarial attacks and defenses against deep neural networks: A survey
• Adversarial Examples: Attacks and Defenses for Deep Learning
• Opportunities and Challenges in Deep Learning Adversarial Robustness: A Survey
• Adversarial Machine Learning in Image Classification: A Survey Towards the Defender’s Perspective
• Adversarial Attacks and Defenses in Images,Graphs and Text A Review

白盒攻击类（也包括可迁移场景）

2017年及以前

• Explaining and Harnessing Adversarial Examples
• Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images
• DeepFool: a simple and accurate method to fool deep neural networks (CVPR 2016)
• The Limitations of Deep Learning in Adversarial Settings (S&P 2016)
• Adversarial Machine Learning at Scale (ICLR 2016)
• Towards Evaluating the Robustness of Neural Networks (S&P 2017)

2017年以后

• Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
• Semantic Adversarial Attacks: Parametric Transformations That Fool Deep Classifiers (ICCV 2019)
• Physical Adversarial Textures That Fool Visual Object Tracking (ICCV 2019)
• advPattern: Physical-World Attacks on Deep Person Re-Identification via Adversarially Transformable Patterns (ICCV 2019)
• Exact Adversarial Attack to Image Captioning via Structured Output Learning with Latent Variables (CVPR 2019)
• Adversarial Attacks Beyond the Image Space (CVPR 2019)
• Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defenses (CVPR 2019)
• Trust Region Based Adversarial Attack on Neural Networks (CVPR 2019)
• Functional Adversarial Attacks (NIPS 2019)
• Cross-Modal Learning with Adversarial Samples (NIPS 2019)
• Adversarial camera stickers: A physical camera-based attack on deep learning systems (ICML 2019)
• Wasserstein Adversarial Examples via Projected Sinkhorn Iterations (ICML 2019)
• Adversarial T-shirt! Evading Person Detectors in A Physical World (ECCV 2020)
• AdvPC: Transferable Adversarial Perturbations on 3D Point Clouds (ECCV 2020)
• Sparse Adversarial Attack via Perturbation Factorization (ECCV 2020)
• Improving the Transferability of Adversarial Examples with Resized-Diverse-Inputs, Diversity-Ensemble and Region Fitting (ECCV 2020)
• Backpropagating Linearly Improves Transferability of Adversarial Examples (NIPS 2020)
• On Adaptive Attacks to Adversarial Example Defenses (NIPS 2020)
• Targeted Adversarial Perturbations for Monocular Depth Prediction (NIPS 2020)
• GreedyFool: Distortion-Aware Sparse Adversarial Attack (NIPS 2020)
• Practical No-box Adversarial Attacks against DNNs (NIPS 2020)
• Stronger and Faster Wasserstein Adversarial Attacks (ICML 2020)
• Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attack (ICML 2020)
• Fooling Detection Alone is Not Enough: Adversarial Attack against Multiple Object Tracking (ICLR 2020)
• Unrestricted Adversarial Examples via Semantic Manipulation (ICLR 2020)
• Learning Transferable Adversarial Perturbations (NIPS 2021)
• Adversarial Attack Generation Empowered by Min-Max Optimization (NIPS 2021)
• Distilling Robust and Non-Robust Features in Adversarial Examples by Information Bottleneck (NIPS 2021)
• Adversarial Robustness with Non-uniform Perturbations (NIPS 2021)
• Fast Minimum-norm Adversarial Attacks through Adaptive Norm Constraints (NIPS 2021)
• Mind the Box: l1-APGD for Sparse Adversarial Attacks on Image Classifiers (ICML 2021)
• Sparse and Imperceptible Adversarial Attack via a Homotopy Algorithm (ICML 2021)
• A Unified Approach to Interpreting and Boosting Adversarial Transferability (ICLR 2021)
• Frequency-driven Imperceptible Adversarial Attack on Semantic Similarity (CVPR 2022)
• [ ]

黑盒攻击类

2017年及以前

• Practical Black-Box Attacks against Machine Learning （CCS 2017）
• UPSET and ANGRI : Breaking High Performance Image Classifiers (arxiv preprint 2017)
• Houdini: Fooling deep structured prediction models (arxiv preprint 2017)

2017年以后

• Boosting Adversarial Attacks with Momentum
• Transferable Perturbations of Deep Feature Distributions
• Dual-Path Distillation: A Unified Framework to Improve Black-Box Attacks
• On the Design of Black-Box Adversarial Examples by Leveraging Gradient-Free Optimization and Operator Splitting Method (ICCV 2019)
• Universal Adversarial Perturbation via Prior Driven Uncertainty Approximation (ICCV 2019)
• Sparse and Imperceivable Adversarial Attacks (ICCV 2019)
• Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks (ICCV 2019)
• Enhancing Adversarial Example Transferability With an Intermediate Level Attack (ICCV 2019)
• The LogBarrier adversarial attack: making effective use of decision boundary information (ICCV 2019)
• Guessing Smart: Biased Sampling for Efficient Black-Box Adversarial Attacks (ICCV 2019)
• Targeted Mismatch Adversarial Attack: Query With a Flower to Retrieve the Tower (ICCV 2019)
• Improving Transferability of Adversarial Examples with Input Diversity (CVPR 2019)
• Feature Space Perturbations Yield More Transferable Adversarial Examples (CVPR 2019)
• Efficient Decision-Based Black-Box Adversarial Attacks on Face Recognition (CVPR 2019)
• Catastrophic Child’s Play: Easy to Perform, Hard to Defend Adversarial Attacks (CVPR 2019)
• Improving Black-box Adversarial Attacks with a Transfer-based Prior (NIPS 2019)
• Simple Black-box Adversarial Attacks (ICML 2019)
• NATTACK: Learning the Distributions of Adversarial Examples for an Improved Black-Box Attack on Deep Neural Networks (ICML 2019)
• Parsimonious Black-Box Adversarial Attacks via Efficient Combinatorial Optimization (ICML 2019)
• Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors (ECCV 2020)
• Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses (ECCV 2020)
• Bias-based Universal Adversarial Patch Attack for Automatic Check-out (ECCV 2020)
• SemanticAdv: Generating Adversarial Examples via Attribute-conditioned Image Editing (ECCV 2020)
• Boosting Decision-based Black-box Adversarial Attacks with Random Sign Flip (ECCV 2020)
• Design and Interpretation of Universal Adversarial Patches in Face Detection (ECCV 2020)
• Motion-Excited Sampler: Video Adversarial Attack with Sparked Prior (ECCV 2020)
• Square Attack: a query-efficient black-box adversarial attack via random search (ECCV 2020)
• Improving Query Efficiency of Black-box Adversarial Attack (ECCV 2020)
• Efficient Adversarial Attacks for Visual Object Tracking (ECCV 2020)
• AdvFlow: Inconspicuous Black-box Adversarial Attacks using Normalizing Flows (NIPS 2020)
• Black-Box Adversarial Attack with Transferable Model-based Embedding (ICLR 2020)
• Sign-OPT: A Query-Efficient Hard-label Adversarial Attack (ICLR 2020)
• BayesOpt Adversarial Attack (ICLR 2020)
• BREAKING CERTIFIED DEFENSES: SEMANTIC ADVERSARIAL EXAMPLES WITH SPOOFED ROBUSTNESS CERTIFICATES (ICLR 2020)
• IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking (CVPR 2021)
• BASAR:Black-box Attack on Skeletal Action Recognition (CVPR 2021)
• Enhancing the Transferability of Adversarial Attacks through Variance Tuning (CVPR 2021)
• Adversarial Attacks on Black Box Video Classifiers: Leveraging the Power of Geometric Transformations (NIPS 2021)
• Meta-Learning the Search Distribution of Black-Box Random Search Based Adversarial Attacks (NIPS 2021)
• Policy-Driven Attack: Learning to Query for Hard-label Black-box Adversarial Examples (ICLR 2021)
• LowKey: Leveraging Adversarial Attacks to Protect Social Media Users from Facial Recognition (ICLR 2021)
• Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon (CVPR 2022)
• Adversarial Texture for Fooling Person Detectors in the Physical World (CVPR 2022)

防御以及鲁棒类

2017年及以前

• Intriguing properties of neural networks
• Adversarial Machine Learning at Scale

2017年以后

• Ensemble Adversarial Training: Attacks and Defenses
• Boosting Adversarial Attacks with Momentum
• Mitigating Adversarial Effects Through Randomization
• Thermometer Encoding: One Hot Way To Resist Adversarial Examples
• Countering Adversarial Images using Input Transformations
• Stochastic activation pruning for robust adversarial defense
• PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
• Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Models
• Towards Deep Learning Models Resistant to Adversarial Attacks (ICLR 2018)
• Adversarial Robustness vs. Model Compression, or Both? (ICCV 2019)
• Evaluating Robustness of Deep Image Super-Resolution Against Adversarial Attacks (ICCV 2019)
• Towards Adversarially Robust Object Detection (ICCV 2019)
• Adversarial Defense via Learning to Generate Diverse Attacks (ICCV 2019)
• Hilbert-Based Generative Defense for Adversarial Examples (ICCV 2019)
• Improving Adversarial Robustness via Guided Complement Entropy (ICCV 2019)
• Defending Against Universal Perturbations With Shared Adversarial Training (ICCV 2019)
• Bilateral Adversarial Training: Towards Fast Training of More Robust Models Against Adversarial Attacks (ICCV 2019)
• CIIDefence: Defeating Adversarial Attacks by Fusing Class-Specific Image Inpainting and Image Denoising (ICCV 2019)
• Feature Denoising for Improving Adversarial Robustness (CVPR 2019)
• Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attack (CVPR 2019)
• Parametric Noise Injection: Trainable Randomness to Improve Deep Neural Network Robustness Against Adversarial Attack (CVPR 2019)
• Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examples (CVPR 2019)
• Searching for a Robust Neural Architecture in Four GPU Hours (CVPR 2019)
• What Does It Mean to Learn in Deep Networks? And, How Does One Detect Adversarial Attacks? (CVPR 2019)
• Detection Based Defense Against Adversarial Examples From the Steganalysis Point of View (CVPR 2019)
• Adversarial Defense Through Network Profiling Based Path Extraction (CVPR 2019)
• ComDefend: An Efficient Image Compression Model to Defend Adversarial Examples (CVPR 2019)
• Curls & Whey: Boosting Black-Box Adversarial Attacks (CVPR 2019)
• Barrage of Random Transforms for Adversarially Robust Defense (CVPR 2019)
• Disentangling Adversarial Robustness and Generalization (CVPR 2019)
• ShieldNets: Defending Against Adversarial Attacks Using Probabilistic Adversarial Robustness (CVPR 2019)
• Defense Against Adversarial Images Using Web-Scale Nearest-Neighbor Search (CVPR 2019)
• Adversarial Defense by Stratified Convolutional Sparse Coding (CVPR 2019)
• Adversarial Defense by Stratified Convolutional Sparse Coding (CVPR 2019)
• Robustness of 3D Deep Learning in an Adversarial Setting (CVPR 2019)
• Defending Against Adversarial Attacks by Randomized Diversification (CVPR 2019)
• Lower Bounds on Adversarial Robustness from Optimal Transport (NIPS 2019)
• Adversarial Robustness through Local Linearization (NIPS 2019)
• On Robustness to Adversarial Examples and Polynomial Optimization (NIPS 2019)
• Model Compression with Adversarial Robustness: A Unified Optimization Framework (NIPS 2019)
• Unlabeled Data Improves Adversarial Robustness (NIPS 2019)
• Theoretical evidence for adversarial robustness through randomization (NIPS 2019)
• Provably robust boosted decision stumps and trees against adversarial attacks (NIPS 2019)
• Robustness to Adversarial Perturbations in Learning from Incomplete Data (NIPS 2019)
• Reverse KL-Divergence Training of Prior Networks: Improved Uncertainty and Adversarial Robustness (NIPS 2019)
• Provable Certificates for Adversarial Examples: Fitting a Ball in the Union of Polytopes (NIPS 2019)
• Are Labels Required for Improving Adversarial Robustness? (NIPS 2019)
• Metric Learning for Adversarial Robustness (NIPS 2019)
• A New Defense Against Adversarial Images: Turning a Weakness into a Strength (NIPS 2019)
• Error Correcting Output Codes Improve Probability Estimation and Adversarial Robustness of Deep Neural Networks (NIPS 2019)
• Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Training (NIPS 2019)
• Tight Certificates of Adversarial Robustness for Randomly Smoothed Classifiers (NIPS 2019)
• Adversarial examples from computational constraints (ICML 2019)
• Certified Adversarial Robustness via Randomized Smoothing (ICML 2019)
• Generalized No Free Lunch Theorem for Adversarial Robustness (ICML 2019)
• On the Connection Between Adversarial Robustness and Saliency Map Interpretability (ICML 2019)
• Adversarial Examples Are a Natural Consequence of Test Error in Noise (ICML 2019)
• Are Generative Classifiers More Robust to Adversarial Attacks? (ICML 2019)
• On Certifying Non-Uniform Bounds against Adversarial Attacks (ICML 2019)
• Improving Adversarial Robustness via Promoting Ensemble Diversity (ICML 2019)
• The Odds are Odd: A Statistical Test for Detecting Adversarial Examples (ICML 2019)
• First-Order Adversarial Vulnerability of Neural Networks and Input Dimension (ICML 2019)
• ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation (ICML 2019)
• Rademacher Complexity for Adversarially Robust Generalization (ICML 2019)
• Robust Design of Deep Neural Networks against Adversarial Attacks based on Lyapunov Theory (CVPR 2020)
• Multitask Learning Strengthens Adversarial Robustness (ECCV 2020)
• Towards Automated Testing and Robustification by Semantic Adversarial Data Generation (ECCV 2020)
• Robust Tracking against Adversarial Attacks (ECCV 2020)
• Connecting the Dots: Detecting Adversarial Perturbations Using Context Inconsistency (ECCV 2020)
• Adversarial Robustness on In- and Out-Distribution Improves Explainability (ECCV 2020)
• Improving Adversarial Robustness by Enforcing Local and Global Compactness (ECCV 2020)
• Defense Against Adversarial Attacks via Controlling Gradient Leaking on Embedded Manifolds (ECCV 2020)
• Inherent Adversarial Robustness of Deep Spiking Neural Networks: Effects of Discrete Input Encoding and Non-Linear Activations (ECCV 2020)
• Manifold Projection for Adversarial Defense on Face Recognition (ECCV 2020)
• Adversarial Robustness of Supervised Sparse Coding (NIPS 2020)
• Biologically Inspired Mechanisms for Adversarial Robustness (NIPS 2020)
• Dual Manifold Adversarial Robustness: Defense against Lp and non-Lp Adversarial Attacks (NIPS 2020)
• On the Tightness of Semidefinite Relaxations for Certifying Robustness to Adversarial Examples (NIPS 2020)
• Adversarial robustness via robust low rank representations (NIPS 2020)
• Fast Adversarial Robustness Certification of Nearest Prototype Classifiers for Arbitrary Seminorms (NIPS 2020)
• Contrastive Learning with Adversarial Examples (NIPS 2020)
• Guided Adversarial Attack for Evaluating and Enhancing Adversarial Defenses (NIPS 2020)
• Black-box Certification and Learning under Adversarial Perturbations (ICML 2020)
• Proper Network Interpretability Helps Adversarial Robustness in Classification (ICML 2020)
• Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks (ICML 2020)
• Sharp Statistical Guaratees for Adversarially Robust Gaussian Classification (ICML 2020)
• Hierarchical Verification for Adversarial Robustness (ICML 2020)
• Implicit Euler Skip Connections: Enhancing Adversarial Robustness via Numerical Stability (ICML 2020)
• Adversarial Robustness Against the Union of Multiple Perturbation Models (ICML 2020)
• Efficiently Learning Adversarially Robust Halfspaces with Noise (ICML 2020)
• Randomization matters How to defend against strong adversarial attacks (ICML 2020)
• Second-Order Provable Defenses against Adversarial Attacks (ICML 2020)
• Overfitting in adversarially robust deep learning (ICML 2020)
• Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial Perturbations (ICML 2020)
• Neural Network Control Policy Verification With Persistent Adversarial Perturbation (ICML 2020)
• Towards Understanding the Regularization of Adversarial Robustness on Neural Networks (ICML 2020)
• Adversarial Robustness via Runtime Masking and Cleansing (ICML 2020)
• Enhancing Transformation-Based Defenses Against Adversarial Attacks with a Distribution Classifier (ICLR 2020)
• Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attacks (ICLR 2020)
• Rethinking Softmax Cross-Entropy Loss for Adversarial Robustness (ICLR 2020)
• Improving Adversarial Robustness Requires Revisiting Misclassified Examples (ICLR 2020)
• Detecting and Diagnosing Adversarial Images with Class-Conditional Capsule Reconstructions (ICLR 2020)
• GAT: Generative Adversarial Training for Adversarial Example Detection and Robust Classification (ICLR 2020)
• Certified Robustness for Top-k Predictions against Adversarial Perturbations via Randomized Smoothing (ICLR 2020)
• Adversarially Robust Representations with Smooth Encoders (ICLR 2020)
• Adversarially robust transfer learning (ICLR 2020)
• Biologically inspired sleep algorithm for increased generalization and adversarial robustness in deep neural networks (ICLR 2020)
• Jacobian Adversarially Regularized Networks for Robustness (ICLR 2020)
• Certified Defenses for Adversarial Patches (ICLR 2020)
• Nesterov Accelerated Gradient and Scale Invariance for Adversarial Attacks (ICLR 2020)
• Provable robustness against all adversarial lp-perturbations for p≥1 (ICLR 2020)
• EMPIR: Ensembles of Mixed Precision Deep Networks for Increased Robustness Against Adversarial Attacks(ICLR 2020)
• Enhancing Transformation-Based Defenses Against Adversarial Attacks with a Distribution Classifier (ICLR 2020)
• Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentation (CVPR 2021)
• When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarks (CVPR 2021)
• Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attack (CVPR 2021)
• LAFEAT: Piercing Through Adversarial Defenses with Latent Features (CVPR 2021)
• LiBRe: A Practical Bayesian Approach to Adversarial Detection (CVPR 2021)
• Zero-shot Adversarial Quantization (CVPR 2021)
• VideoMoCo: Contrastive Video Representation Learning with Temporally Adversarial Examples (CVPR 2021)
• Adversarial Robustness under Long-Tailed Distribution (CVPR 2021)
• Renofeation: A Simple Transfer Learning Method for Improved Adversarial Robustness (CVPR 2021)
• AugMax: Adversarial Composition of Random Augmentations for Robust Training (NIPS 2021)
• Revisiting Hilbert-Schmidt Information Bottleneck for Adversarial Robustness (NIPS 2021)
• Towards Better Understanding of Training Certifiably Robust Models against Adversarial Examples (NIPS 2021)
• Shift Invariance Can Reduce Adversarial Robustness (NIPS 2021)
• Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networks (NIPS 2021)
• Adversarial Robustness with Semi-Infinite Constrained Learning (NIPS 2021)
• Do Wider Neural Networks Really Help Adversarial Robustness? (NIPS 2021)
• Adversarial Examples in Multi-Layer Random ReLU Networks (NIPS 2021)
• Generalized Depthwise-Separable Convolutions for Adversarially Robust and Efficient Neural Networks (NIPS 2021)
• Stable Neural ODE with Lyapunov-Stable Equilibrium Points for Defending Against Adversarial Attacks (NIPS 2021)
• Exponential Separation between Two Learning Models and Adversarial Robustness (NIPS 2021)
• When does Contrastive Learning Preserve Adversarial Robustness from Pretraining to Finetuning? (NIPS 2021)
• Automated Discovery of Adaptive Attacks on Adversarial Defenses (NIPS 2021)
• Can we have it all? On the Trade-off between Spatial and Adversarial Robustness of Neural Networks (NIPS 2021)
• Encoding Robustness to Image Style via Adversarial Feature Perturbations (NIPS 2021)
• Adversarially robust learning for security-constrained optimal power flow (NIPS 2021)
• Neural Architecture Dilation for Adversarial Robustness (NIPS 2021)
• Clustering Effect of Adversarial Robust Models (NIPS 2021)
• CARTL: Cooperative Adversarially-Robust Transfer Learning (ICML 2021)
• SPADE: A Spectral Method for Black-Box Adversarial Robustness Evaluation (ICML 2021)
• Adversarial Robustness Guarantees for Random Deep Neural Networks (ICML 2021)
• Learning Diverse-Structured Networks for Adversarial Robustness (ICML 2021)
• Weight-covariance alignment for adversarially robust neural networks (ICML 2021)
• Maximum Mean Discrepancy Test is Aware of Adversarial Attacks (ICML 2021)
• Knowledge Enhanced Machine Learning Pipeline against Diverse Adversarial Attacks (ICML 2021)
• CIFS: Improving Adversarial Robustness of CNNs via Channel-wise Importance-based Feature Selection (ICML 2021)
• Towards Defending against Adversarial Examples via Attack-Invariant Features (ICML 2021)
• Improving Adversarial Robustness via Channel-wise Activation Suppressing (ICLR 2021)
• Improving VAEs' Robustness to Adversarial Attack (ICLR 2021)
• Heating up decision boundaries: isocapacitory saturation, adversarial scenarios and generalization bounds (ICLR 2021)
• Perceptual Adversarial Robustness: Defense Against Unseen Threat Models (ICLR 2021)
• Self-supervised Adversarial Robustness for the Low-label, High-data Regime (ICLR 2021)
• Provably robust classification of adversarial examples with detection (ICLR 2021)
• Stochastic Security: Adversarial Defense Using Long-Run Dynamics of Energy-Based Models (ICLR 2021)
• ARMOURED: Adversarially Robust MOdels using Unlabeled data by REgularizing Diversity (ICLR 2021)
• Improving the Transferability of Targeted Adversarial Examples through Object-Based Diverse Input (CVPR 2022)
• Enhancing Adversarial Training with Second-Order Statistics of Weights (CVPR 2022)
• Practical Evaluation of Adversarial Robustness via Adaptive Auto Attack (CVPR 2022)
• On Adversarial Robustness of Trajectory Prediction for Autonomous Vehicles (CVPR 2022)
• Enhancing Adversarial Robustness for Deep Metric Learning (CVPR 2022)
• [ ]

其他

• Detecting Overfitting via Adversarial Examples (NIPS 2019)
• On Relating Explanations and Adversarial Examples (NIPS 2019)
• Cross-Domain Transferability of Adversarial Perturbations (NIPS 2019)
• Adversarial Examples Are Not Bugs, They Are Features (NIPS 2019)
• A Game Theoretic Analysis of Additive Adversarial Attacks and Defenses (NIPS 2020)
• Most ReLU Networks Suffer from ℓ2ℓ2 Adversarial Perturbations (NIPS 2020)
• On the Trade-off between Adversarial and Backdoor Robustness (NIPS 2020)
• HYDRA: Pruning Adversarially Robust Neural Networks (NIPS 2020)
• APRICOT: A Dataset of Physical Adversarial Attacks on Object Detection (ECCV 2020)
• Towards a Unified Game-Theoretic View of Adversarial Perturbations and Robustness (NIPS 2021)
• A PAC-Bayes Analysis of Adversarial Robustness (NIPS 2021)
• Adversarial Examples Make Strong Poisons (NIPS 2021)
• Query Complexity of Adversarial Attacks (ICML 2021)
• Mixed Nash Equilibria in the Adversarial Examples Game (ICML 2021)