Malware Classification Using Diluted Convolutional Neural Network with Fast Gradient Sign Method

2025-10-17

2025 2nd International Conference on Software, Systems and Information Technology (SSITCON)

AI Summary

This study addresses the growing complexity and stealthiness of Android malware, which undermine the efficiency of traditional detection methods and their reliance on high-dimensional features. To this end, the authors propose a lightweight adversarial training framework that integrates Dilated Convolutional Neural Networks (DICNN) with the Fast Gradient Sign Method (FGSM). By leveraging dilated convolutions to expand the receptive field, the model effectively captures long-range, dispersed malicious patterns. During training, a single-step FGSM perturbation is introduced to enhance robustness and classification accuracy without increasing model parameter count or dependence on high-dimensional features. Experimental results on public datasets demonstrate that the proposed model achieves a detection accuracy of 99.44%, significantly outperforming existing approaches such as standard DCNN.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Android malware has become an increasingly critical threat to organizations, society and individuals, posing significant risks to privacy, data security and infrastructure. As malware continues to evolve in terms of complexity and sophistication, the mitigation and detection of these malicious software instances have become more time consuming and challenging particularly due to the requirement of large number of features to identify potential malware. To address these challenges, this research proposes Fast Gradient Sign Method with Diluted Convolutional Neural Network (FGSM -DICNN) method for malware classification. DICNN contains diluted convolutions which increases receptive field, enabling the model to capture dispersed malware patterns across long ranges using fewer features without adding parameters. Additionally, the FGSM strategy enhance the accuracy by using one-step perturbations during training that provides more defensive advantage of lower computational cost. This integration helps to manage high classification accuracy while reducing the dependence on extensive feature sets. The proposed FGSMDICNN model attains 99.44% accuracy while outperforming other existing approaches such as Custom Deep Neural Network (DCNN).

Problem

• Android malware
• malware classification
• feature dependency
• malware detection
• malware complexity

Innovation

• Diluted Convolution
• Fast Gradient Sign Method
• Malware Classification
• Adversarial Training
• Receptive Field Expansion