Electronics

https://www.mdpi.com/journal/electronics/special_issues/E3Q62WWT2E

Machine Learning (ML) has revolutionized cyber security and privacy by enabling advanced threat detection, anomaly identification, and automated defense mechanisms. From intrusion detection systems to privacy-preserving data analytics, ML-driven solutions are increasingly embedded in critical infrastructures, IoT ecosystems, and cloud-based services. However, the rapid adoption of ML technologies has also exposed vulnerabilities that malicious actors exploit, such as adversarial attacks on ML models, data poisoning, membership inference attacks, and model inversion attacks. Furthermore, privacy concerns, especially in federated learning and generative AI, raise ethical and regulatory challenges that demand urgent attention.

This Special Issue will address the dual role of ML in cyber security and privacy as both a tool for defense and a vector for attack. We invite cutting-edge research that explores novel threats, develops robust mitigation strategies, and establishes ethical frameworks for deploying ML in sensitive domains. Submissions should emphasize interdisciplinary approaches, bridging ML theory, cryptographic techniques, policy design, and real-world applications.

We welcome original research articles, comprehensive reviews, and case studies focused on (but not limited to) the following themes:

1. ML-Driven Threat Detection and Mitigation

• Novel ML methods for identifying zero-day exploits, ransomware, and APTs (Advanced Persistent Threats);
• Adversarial robustness in malware classification, network intrusion detection, and phishing detection systems;
• Explainable AI (XAI) for transparent threat analysis and incident response.

2. Privacy-Preserving ML in Sensitive Domains

• Federated learning architectures for secure data collaboration in healthcare, finance, and smart cities;
• Differential privacy guarantees in ML training and inference;
• Mitigating model inversion and membership inference attacks in generative models (e.g., GANs, LLMs).

3. Attacks on ML Systems

• Adversarial attacks targeting real-time decision systems (e.g., autonomous vehicles, critical infrastructure);
• Data poisoning in federated learning and edge computing environments;
• Privacy breaches via model extraction or side-channel attacks.

4. Ethical and Regulatory Challenges

• Bias and fairness in ML-based security systems (e.g., facial recognition, predictive policing);
• Compliance with GDPR, CCPA, and other privacy regulations in ML deployments;
• Human-in-the-loop frameworks for accountable security automation.

5. Emerging Applications and Case Studies

• ML for securing blockchain networks and decentralized applications;
• Quantum-resistant ML algorithms for post-quantum cryptography;
• Real-world deployments in industrial control systems (ICS), 5G networks, and IoT ecosystems.