Machine Learning and Federated Learning in Industrial Cybersecurity
Rampelli Manojkumar, Sanivarapu Prasanth Vaidya, Prasanta Kumar Jena, S. Ashok, Sandeep Dasari · Advances in computational intelligence and robotics book series · 2025
The integration of Artificial Intelligence and Machine Learning into industrial cybersecurity addresses the limitations of traditional methods against evolving cyber threats. This chapter explores supervised, unsupervised, and reinforcement learning techniques for threat detection, anomaly detection, and adaptive response mechanisms. It emphasizes Federated Learning (FL) as a privacy-preserving approach in the Industrial Internet of Things, enabling collaborative model training without data exposure. Ethical and legal challenges, including bias, accountability, and regulatory compliance, are analyzed to ensure responsible AI deployment. Case studies in automotive, energy, and industrial control systems demonstrate FL's effectiveness in predictive maintenance. Future trends such as Quantum AI and Explainable AI are discussed for enhancing cryptographic security and transparency. The chapter concludes with managerial implications for adopting AI-driven solutions, emphasizing privacy, and cross-industry collaboration to safeguard critical infrastructure in Industry 5.0.