Technique Talk: Federated Learning in Wireless Networks: A Security Perspective
Federated learning in wireless networks represents a significant advancement in the field of distributed artificial intelligence (AI). This approach allows machine learning models to be trained directly on users’ devices, such as phones, autonomous vehicles, or drones, without the need to transfer locally collected data to a central server. From the security perspective, federated learning can be used to detect cyberattacks in wireless networks. In this context, we will discuss an initial solution that leverages distributed federated learning to detect attacks. However, federated learning can itself be prone to cyberattacks. Consequently, we will explore reliable unmanned aerial vehicle (UAV) client participation in federated learning under attack conditions. Together, these studies highlight the importance of the interplay between federated learning and security for wireless networks.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- Maxwell Theatre
- Trinity College Dublin
- Dublin, Dublin
- Ireland
- Building: Hamilton Building
Speakers
Prof. Wael Jaafar of École de Technologie Supérieure (ÉTS), University of Quebec, Montreal
Biography:
Wael Jaafar (S’08, M’14, SM’20) is an Associate Professor at the Software and IT Engineering Department of École de Technologie Supérieure (ÉTS), University of Quebec, Montreal, Canada since September 2022. He holds Master’s and PhD degrees from Polytechnique Montreal, Canada. Between 2019 and 2022, Dr. Jaafar was an NSERC Postdoctoral Fellow with the Systems and Computer Engineering Department of Carleton University. From 2014 to 2018, he pursued a career in the telecommunications industry where he has been involved in designing telecommunication solutions for projects across Canada and abroad. During his career, Dr. Jaafar was a visiting researcher at Khalifa University, Abu Dhabi, UAE in 2019, Keio University, Tokyo, Japan in 2013, and UQAM, Montreal, Canada in 2007. He received several prestigious grants including the National Sciences and Engineering Research Council of Canada (NSERC) Alexander-Graham Bell scholarship, the Fonds de recherche du Québec–Nature et technologies (FRQNT) scholarship, and best paper awards at IEEE ICC 2021, ISCC 2023, and CIoT 2024. His research interests include wireless communications, integrated terrestrial and non-terrestrial networks, resource allocation, edge caching and computing, machine learning for communication and networks, and cybersecurity.