Keeping the Bad Guys Out: Protecting and Vaccinating Deep Learning with JPEG Compression
This addresses a critical security problem for AI systems in image recognition, offering a practical defense against adversarial attacks, though it is incremental as it builds on existing compression techniques.
The paper tackles the vulnerability of deep neural networks to adversarial attacks by proposing JPEG compression as a pre-processing step to reduce their effects, demonstrating that it can dramatically lower attack success rates, such as reducing misclassification from 90% to 10% in some cases.
Deep neural networks (DNNs) have achieved great success in solving a variety of machine learning (ML) problems, especially in the domain of image recognition. However, recent research showed that DNNs can be highly vulnerable to adversarially generated instances, which look seemingly normal to human observers, but completely confuse DNNs. These adversarial samples are crafted by adding small perturbations to normal, benign images. Such perturbations, while imperceptible to the human eye, are picked up by DNNs and cause them to misclassify the manipulated instances with high confidence. In this work, we explore and demonstrate how systematic JPEG compression can work as an effective pre-processing step in the classification pipeline to counter adversarial attacks and dramatically reduce their effects (e.g., Fast Gradient Sign Method, DeepFool). An important component of JPEG compression is its ability to remove high frequency signal components, inside square blocks of an image. Such an operation is equivalent to selective blurring of the image, helping remove additive perturbations. Further, we propose an ensemble-based technique that can be constructed quickly from a given well-performing DNN, and empirically show how such an ensemble that leverages JPEG compression can protect a model from multiple types of adversarial attacks, without requiring knowledge about the model.