PRESENTER
- Yalin Sagduyu, Virginia Tech National Security Institute, Arlington, VA, USA
- Tugba Erpek, Virginia Tech National Security Institute, Arlington, VA, USA
SCOPE
Machine learning has rapidly permeated the field of wireless communications, enabling a myriad of applications including spectrum sensing, RF fingerprinting, spectrum access, resource allocation, and waveform design. However, this proliferation of machine learning has also attracted adversarial threats that can effectively exploit vulnerabilities of machine learning algorithms and disrupt training and test-time operations. This tutorial aims to provide a comprehensive understanding of adversarial machine learning in wireless systems by considering a variety of security and privacy threats. We will highlight the potential impact of these attacks and discuss defense mechanisms for various machine learning techniques, including deep neural networks, reinforcement learning, and federated learning when applied to NextG communications systems. By attending this tutorial, participants are expected to learn about the emerging attack vectors on machine learning -driven wireless systems and corresponding defense approaches that have important implications in multi-disciplinary research and development efforts encompassing wireless communications, networking, and security.
SHORT BIO
Yalin Sagduyu received the B.S. degree in electrical and electronics engineering from Bogazici University, Istanbul, Turkey, and the M.S. and Ph.D. degrees in electrical and computer engineering from the University of Maryland, College Park, MD, USA. He is currently a Research Professor with Virginia Tech National Security Institute, Arlington, VA, USA. Prior to that, he was the Director of networks and security with Intelligent Automation, Inc./BlueHalo, Rockville, MD, USA. He is also a Visiting Research Professor with the Department of Electrical and Computer Engineering, University of Maryland. His research interests include wireless communications, networks, security, and machine learning. He has extensively published in the area of adversarial machine learning for NextG communications systems. He is an Editor of IEEE Transactions on Communications and an Editor of IEEE Transactions on Cognitive Communications and Networking. He chaired workshops at ACM MobiCom, ACM WiSec, IEEE CNS, and IEEE ICNP. He was a Track Chair at IEEE PIMRC, IEEE GlobalSIP, and IEEE MILCOM, and served in the Organizing Committee of IEEE GLOBECOM and IEEE MILCOM. He gave tutorials at IEEE GLOBECOM and IEEE MILCOM. He was the recipient of IEEE HST 2018 Best Paper Award for his paper on machine learning security.
Tugba Erpek is a Research Associate Professor in the Intelligent Systems Division of the Virginia Tech National Security Institute. She received her Ph.D. degree in Electrical and Computer Engineering from Virginia Tech. Prior to joining to Virginia Tech, she was a Lead Scientist and Network Communications Technical Area Lead at the Intelligent Automation, a BlueHalo Company and a Senior Communications Systems Engineer at the Shared Spectrum Company. Her research interests are in wireless communications and networks, 5G and beyond, wireless security, resource allocation, machine learning and adversarial machine learning applications in NextG communication systems. She has published extensively in these areas. She has been serving as a TPC member and reviewer for major IEEE conferences and journals.
