Shang-Tse Chen’s research while affiliated with Georgia Institute of Technology and other places

Deep learning models are being integrated into a wide range of high-impact, security-critical systems, from self-driving cars to medical diagnosis. However, recent research has demonstrated that many of these deep learning architectures are vulnerable to adversarial attacks--highlighting the vital need for defensive techniques to detect and mitigate these attacks before they occur. To combat these adversarial attacks, we developed UnMask, an adversarial detection and defense framework based on robust feature alignment. The core idea behind UnMask is to protect these models by verifying that an image's predicted class ("bird") contains the expected robust features (e.g., beak, wings, eyes). For example, if an image is classified as "bird", but the extracted features are wheel, saddle and frame, the model may be under attack. UnMask detects such attacks and defends the model by rectifying the misclassification, re-classifying the image based on its robust features. Our extensive evaluation shows that UnMask (1) detects up to 96.75% of attacks, with a false positive rate of 9.66% and (2) defends the model by correctly classifying up to 93% of adversarial images produced by the current strongest attack, Projected Gradient Descent, in the gray-box setting. UnMask provides significantly better protection than adversarial training across 8 attack vectors, averaging 31.18% higher accuracy. Our proposed method is architecture agnostic and fast. We open source the code repository and data with this paper: https://github.com/unmaskd/unmask.

In this talk we describe our content-preserving attack on object detectors, ShapeShifter, and demonstrate how to evaluate this threat in realistic scenarios. We describe how we use CARLA, a realistic urban driving simulator, to create these scenarios, and how we use ShapeShifter to generate content-preserving attacks against those scenarios.

Adversarial machine learning research has recently demonstrated the feasibility to confuse automatic speech recognition (ASR) models by introducing acoustically imperceptible perturbations to audio samples. To help researchers and practitioners gain better understanding of the impact of such attacks, and to provide them with tools to help them more easily evaluate and craft strong defenses for their models, we present Adagio, the first tool designed to allow interactive experimentation with adversarial attacks and defenses on an ASR model in real time, both visually and aurally. Adagio incorporates AMR and MP3 audio compression techniques as defenses, which users can interactively apply to attacked audio samples. We show that these techniques, which are based on psychoacoustic principles, effectively eliminate targeted attacks, reducing the attack success rate from 92.5% to 0%. We will demonstrate Adagio and invite the audience to try it on the Mozilla Common Voice dataset. Code related to this paper is available at: https://github.com/nilakshdas/ADAGIO.

Given the ability to directly manipulate image pixels in the digital input space, an adversary can easily generate imperceptible perturbations to fool a Deep Neural Network (DNN) image classifier, as demonstrated in prior work. In this work, we propose ShapeShifter, an attack that tackles the more challenging problem of crafting physical adversarial perturbations to fool image-based object detectors like Faster R-CNN. Attacking an object detector is more difficult than attacking an image classifier, as it needs to mislead the classification results in multiple bounding boxes with different scales. Extending the digital attack to the physical world adds another layer of difficulty, because it requires the perturbation to be robust enough to survive real-world distortions due to different viewing distances and angles, lighting conditions, and camera limitations. We show that the Expectation over Transformation technique, which was originally proposed to enhance the robustness of adversarial perturbations in image classification, can be successfully adapted to the object detection setting. ShapeShifter can generate adversarially perturbed stop signs that are consistently mis-detected by Faster R-CNN as other objects, posing a potential threat to autonomous vehicles and other safety-critical computer vision systems. Code related to this paper is available at: https://github.com/shangtse/robust-physical-attack.

The rapidly growing body of research in adversarial machine learning has demonstrated that deep neural networks (DNNs) are highly vulnerable to adversarially generated images. This underscores the urgent need for practical defense techniques that can be readily deployed to combat attacks in real-time. Observing that many attack strategies aim to perturb image pixels in ways that are visually imperceptible, we place JPEG compression at the core of our proposed SHIELD defense framework, utilizing its capability to effectively "compress away" such pixel manipulation. To immunize a DNN model from artifacts introduced by compression, SHIELD "vaccinates" the model by retraining it with compressed images, where different compression levels are applied to generate multiple vaccinated models that are ultimately used together in an ensemble defense. On top of that, SHIELD adds an additional layer of protection by employing randomization at test time that compresses different regions of an image using random compression levels, making it harder for an adversary to estimate the transformation performed. This novel combination of vaccination, ensembling, and randomization makes SHIELD a fortified multi-pronged defense. We conducted extensive, large-scale experiments using the ImageNet dataset, and show that our approaches eliminate up to 98% of gray-box attacks delivered by strong adversarial techniques such as Carlini-Wagner's L2 attack and DeepFool. Our approaches are fast and work without requiring knowledge about the model.

Adversarial machine learning research has recently demonstrated the feasibility to confuse automatic speech recognition (ASR) models by introducing acoustically imperceptible perturbations to audio samples. To help researchers and practitioners gain better understanding of the impact of such attacks, and to provide them with tools to help them more easily evaluate and craft strong defenses for their models, we present ADAGIO, the first tool designed to allow interactive experimentation with adversarial attacks and defenses on an ASR model in real time, both visually and aurally. ADAGIO incorporates AMR and MP3 audio compression techniques as defenses, which users can interactively apply to attacked audio samples. We show that these techniques, which are based on psychoacoustic principles, effectively eliminate targeted attacks, reducing the attack success rate from 92.5% to 0%. We will demonstrate ADAGIO and invite the audience to try it on the Mozilla Common Voice dataset.

Given the ability to directly manipulate image pixels in the digital input space, an adversary can easily generate imperceptible perturbations to fool a Deep Neural Network (DNN) image classifier, as demonstrated in prior work. In this work, we tackle the more challenging problem of crafting physical adversarial perturbations to fool image-based object detectors like Faster R-CNN. Attacking an object detector is more difficult than attacking an image classifier, as it needs to mislead the classification results in multiple bounding boxes with different scales. Extending the digital attack to the physical world adds another layer of difficulty, because it requires the perturbation to be robust enough to survive real-world distortions due to different viewing distances and angles, lighting conditions, and camera limitations. We show that the Expectation over Transformation technique, which was originally proposed to enhance the robustness of adversarial perturbations in image classification, can be successfully adapted to the object detection setting. Our approach can generate adversarially perturbed stop signs that are consistently mis-detected by Faster R-CNN as other objects, posing a potential threat to autonomous vehicles and other safety-critical computer vision systems.

The rapidly growing body of research in adversarial machine learning has demonstrated that deep neural networks (DNNs) are highly vulnerable to adversarially generated images. This underscores the urgent need for practical defense that can be readily deployed to combat attacks in real-time. Observing that many attack strategies aim to perturb image pixels in ways that are visually imperceptible, we place JPEG compression at the core of our proposed Shield defense framework, utilizing its capability to effectively "compress away" such pixel manipulation. To immunize a DNN model from artifacts introduced by compression, Shield "vaccinates" a model by re-training it with compressed images, where different compression levels are applied to generate multiple vaccinated models that are ultimately used together in an ensemble defense. On top of that, Shield adds an additional layer of protection by employing randomization at test time that compresses different regions of an image using random compression levels, making it harder for an adversary to estimate the transformation performed. This novel combination of vaccination, ensembling, and randomization makes Shield a fortified multi-pronged protection. We conducted extensive, large-scale experiments using the ImageNet dataset, and show that our approaches eliminate up to 94% of black-box attacks and 98% of gray-box attacks delivered by the recent, strongest attacks, such as Carlini-Wagner's L2 and DeepFool. Our approaches are fast and work without requiring knowledge about the model.

Cybersecurity analysts are often presented suspicious machine activity that does not conclusively indicate compromise, resulting in undetected incidents or costly investigations into the most appropriate remediation actions. There are many reasons for this: deficiencies in the number and quality of security products that are deployed, poor configuration of those security products, and incomplete reporting of product-security telemetry. Managed Security Service Providers (MSSP's), which are tasked with detecting security incidents on behalf of multiple customers, are confronted with these data quality issues, but also possess a wealth of cross-product security data that enables innovative solutions. We use MSSP data to develop Virtual Product, which addresses the aforementioned data challenges by predicting what security events would have been triggered by a security product if it had been present. This benefits the analysts by providing more context into existing security incidents (albeit probabilistic) and by making questionable security incidents more conclusive. We achieve up to 99% AUC in predicting the incidents that some products would have detected had they been present.