Analyzing Malware Behavior Using Generative Neural Networks

Abstract

Traditional detection techniques are struggling with ever-evolving malware threats like zero-day attacks, polymorphic malware, and adversarial samples. Current detection systems (signature-based, heuristic-based, conventional machine learning) fail to generalize to unseen/obfuscated malware variants. In an attempt to overcome these constraints, this paper investigates the possibilities of employing Generative Neural Networks (GNNs), in the form of Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) for the purpose of malware be haviour analysis and detection. We aim to create a novel framework for detecting malware samples that provides some of the best performance in terms of accuracy, precision, and recall while remaining robust to new or unseen malware. This work aims to firstly implement a generative learning-based approach and to measure its adversarial robustness in comparison with the four existing detection techniques. Experimental results show the accuracy, precision, recall of the proposed model is found to be 96.5%, 95.9%, 94.6% with the false positive rate of the model which can be negligible and it is 3.2% which outperforms the traditional machine learning and deep learning models. Our results demonstrate that GNN-based malware detection not only addresses the limitations of conventional approaches in terms of scalability but also provides a more robust and adaptable framework that could be integrated into future real-time threat intelligence and automated defense systems.