Lists
Lists
Ghanem, Mohamed Chahine (2024) Advancing IoT and Cloud Security through LLMs, Federated Learning, and Reinforcement Learning. In: 7th IEEE Conference on Cloud and Internet of Things, 29-31 October 2024, Montreal, Canada.
The rapid expansion of the Internet of Things (IoT) and cloud computing has revolutionized numerous industries by enhancing connectivity, automation, and data processing capabilities. However, this growth has also introduced significant security challenges, with IoT devices and cloud infrastructures becoming prime targets for cyberattacks. Recent developments in AI, particularly Large Language Models (LLMs), Federated Learning (FL) and Deep Reinforcement Learning (DRL), present innovative solutions to bolster IoT and cloud security. This talk explores the integration of LLMs and DRL to enhance many security aspects such as communication security, intrusion detection, incident response, forensic investigation as well as enforcing security by design in IoT and cloud environments.
Large Language Models (LLMs) exploit their advanced natural language processing capabilities to scrutinize extensive datasets, thereby identifying anomalies and forecasting potential threats. Simultaneously, Deep Reinforcement Learning (DRL) algorithms enhance security protocols through ongoing learning and adaptation to the evolving threat landscape. The Talk will highlight the synergistic potential of integrating LLMs, FL and DRL to deliver robust, distributed, and real-time security monitoring and automated mitigation strategies. The Integration and orchestration of these approaches could constitute a proactive defence mechanism, markedly enhancing the resilience of IoT and cloud infrastructures against cyberattacks. The talk will also discuss future research directions notably refining these AI models for heightened accuracy and redefining cybersecurity paradigms for IoT and cloud computing.
 |
View Item |