Defense Against Adversarial Attacks using Convolutional Auto-Encoders
This addresses the issue of adversarial robustness for deep learning systems, but it appears incremental as it builds on existing auto-encoder methods.
The paper tackles the problem of deep learning models being vulnerable to adversarial attacks by using convolutional auto-encoders to restore model accuracy, achieving results that effectively counter adversarial perturbations in input images.
Deep learning models, while achieving state-of-the-art performance on many tasks, are susceptible to adversarial attacks that exploit inherent vulnerabilities in their architectures. Adversarial attacks manipulate the input data with imperceptible perturbations, causing the model to misclassify the data or produce erroneous outputs. This work is based on enhancing the robustness of targeted classifier models against adversarial attacks. To achieve this, an convolutional autoencoder-based approach is employed that effectively counters adversarial perturbations introduced to the input images. By generating images closely resembling the input images, the proposed methodology aims to restore the model's accuracy.