UNICAD: A Unified Approach for Attack Detection, Noise Reduction and Novel Class Identification

TL;DR

UNICAD is a comprehensive framework that enhances DNN robustness by simultaneously detecting adversarial attacks, identifying unseen classes, and recovering from attacks, demonstrated effectively on CIFAR-10.

Contribution

It introduces a unified approach combining prototype-based DNNs and autoencoders for attack detection, noise reduction, and novel class identification.

Findings

Outperforms traditional models in adversarial mitigation

Accurately detects unseen classes in image classification

Effectively recovers from adversarial attacks on CIFAR-10

Abstract

As the use of Deep Neural Networks (DNNs) becomes pervasive, their vulnerability to adversarial attacks and limitations in handling unseen classes poses significant challenges. The state-of-the-art offers discrete solutions aimed to tackle individual issues covering specific adversarial attack scenarios, classification or evolving learning. However, real-world systems need to be able to detect and recover from a wide range of adversarial attacks without sacrificing classification accuracy and to flexibly act in {\bf unseen} scenarios. In this paper, UNICAD, is proposed as a novel framework that integrates a variety of techniques to provide an adaptive solution. For the targeted image classification, UNICAD achieves accurate image classification, detects unseen classes, and recovers from adversarial attacks using Prototype and Similarity-based DNNs with denoising autoencoders. Our experiments performed on the CIFAR-10 dataset highlight UNICAD's effectiveness in adversarial mitigation and unseen class classification, outperforming traditional models.