Android Malware Detection

Machine learning for the detection of Android malware

In the field of malware detection and mitigation, my research is focused on countering increasingly sophisticated, evasive, and dynamic attack vectors through the development of machine learning and deep learning-based detection systems. The operational context encompasses both the protection of critical infrastructure and the security of mobile devices and IoT networks—areas where manual analysis of traditional signatures has become ineffective.

To address these challenges, advanced methodological solutions have been designed and implemented based on the integration of deep learning techniques and hybrid analysis (Augello et al., 2025). By combining static code analysis with dynamic tracking of runtime behavior, AI models have been developed that are capable of autonomously classifying complex malicious patterns and extracting identifying features even from enormous volumes of unstructured data.

One of the solutions explored against model aging is the adoption of Federated Learning (Augello et al., 2025). This paradigm allows interconnected nodes to cooperatively train shared detection models while keeping local data private, enabling crowdsourced learning and adaptation to evolving threats without the need for expensive centralized data collection. To ensure long-term sustainability and participation in crowdsourced mobile security networks, I have also explored economic and game-theoretic models, such as Stackelberg game frameworks (Augello et al., 2025). These models incentivize mobile users to share local computing resources based on their capabilities and constraints, ensuring a balanced contribution scheme.

Ongoing research efforts are focused on scarce label scenarios, where the availability of labeled malware samples is limited; adversarial robustness and explainability of AI models (Augello et al., 2026); and the development of lightweight detection systems suitable for resource-constrained devices.

References

2026

  1. Resource-Aware Federated Learning for Malware Detection on Smart Devices
    Andrea Augello, Alessandra De Paola, and Giuseppe Lo Re
    In 2026 IEEE International Conference on Smart Computing (SmartComp), 2026

2025

  1. Hybrid Multilevel Detection of Mobile Devices Malware Under Concept Drift
    Andrea Augello, Alessandra De Paola, and Giuseppe Lo Re
    Journal of Network and Systems Management, Feb 2025
  2. M2FD: Mobile malware federated detection under concept drift
    Andrea Augello, Alessandra De Paola, and Giuseppe Lo Re
    Computers & Security, 2025
  3. A Stackelberg Approach to Federated Learning for Malware Detection
    Andrea Augello, Alessandra De Paola, Marena Jestin, and 1 more author
    In CEUR Workshop Proceedings - Joint National Conference on Cybersecurity, ITASEC & SERICS, 2025