T3: Deep Learning for Network Optimization and Resource Allocation in 6G and Beyond
Organizer: Hina Tabassum, York University, Canada
Organizer: Aryan Kaushik, University of Sussex, UK
Organizer: Carlo Fischione, KTH, Royal Institute of Technology, Sweden
Abstract: The next generation of wireless networks is anticipated to be more complex and heterogeneous due to higher transmission frequencies, massive internet-of-things (IoT) devices in air-space-ground networks, and ultra-dense access points. Subsequently, the wireless channel coherence time is reducing which necessitates faster and proactive resource management. Machine learning (ML), specifically deep learning, has shown promises to overcome those challenges by training neural networks (NNs) in an offline manner. Once trained, the time complexity of obtaining network resource allocation variables from the NNs become significantly lower than the traditional optimization-based approaches. In the sequel, this tutorial will first provide an overview of state-of-the-art machine learning solutions that can reduce the time complexity of resource allocation in future wireless networks. As well, the aspects related to both “ML-assisted wireless solutions” and “Wireless-enabled ML services” will be covered. We will elaborate on the fundamental concepts, challenges, and applications related of supervised, unsupervised, and reinforcement learning for 6G network resource allocation. Next, the tutorial will delve into the deep unsupervised learning methods for network resource allocation problems with non-linear and non-convex constraints. The use of implicit layers and differentiable projection methods will be discussed. The tutorial will then focus on case-studies demonstrating the applications of reinforcement learning in vehicular networks, IoT, and non-terrestrial networks (NTNs). Finally, the tutorial will cover the significance of the concepts related to centralized and distributed learning as well as ‘over-the-air’ federated Learning in 6G and beyond. The tutorial will conclude by pointing out the existing research gaps in the successful roll-out of ML-enabled resource allocation and highlight potential research directions.
Organizer’s bios:
Prof. Hina Tabassum (SM’17) is currently an Associate Professor at the Lassonde School of Engineering, York University, Canada, where she joined as an Assistant Professor in 2018. She is also appointed as York Research Chair on 5G/6Genabled mobility and sensing applications in 2023 for five years. She received her PhD degree from King Abdullah University of Science and Technology (KAUST) in 2013, and completed postdoctoral research at University of Manitoba, Canada, in 2018. Dr. Tabassum has made significant contributions in the development of a diverse spectrum of statistical models and numerical optimization algorithms. These innovations are tailored to enhance the performance of 5G and 6G wireless networks, catering to a wide array of applications encompassing vehicular, aerial, space, and IoT sensing networks. She received Lassonde Innovation Early-Career Researcher Award in 2023, N2Women: Rising Stars in Computer Networking and Communications in 2022, and listed in the Stanford’s list of the World’s Top Two-Percent Researchers in 2021, 2022, and 2023. She has published over 90 refereed articles in well-reputed IEEE journals, magazines, and conferences (https://sites.google.com/a/kaust.edu.sa/hina-tabassum/). Her publications thus far have garnered 5500+ citations with an h-index of 34 (according to Google Scholar). She delivered several tutorials and invited talks, including recently IEEE PIMRC’22, IEEE WCNC’23, IEEE IoT World Forum’23.
Prof. Aryan Kaushik is Assistant Professor at the University of Sussex, UK, since 2021. He has been with University College London, UK, University of Edinburgh, UK, and Hong Kong University of Science and Technology, Hong Kong. He has held visiting appointments at Imperial College London, UK, University of Luxembourg, Luxembourg, Athena RC, Greece, and Beihang University, China. He has been a panellist for the UKRI EPSRC ICT Prioritisation Panel 2023, Editor of two upcoming books on Integrated Sensing and Communications, and Non-Terrestrial Networks to be published by Elsevier, and PhD External Examiner internationally such as at UC3M, Spain. He has been Editor for IEEE Communications Technology News, IEEE OJCOMS, IEEE Communications Letters, Guest Editor for IEEE IoT Magazine, IEEE OJCOMS, and many others, Invited Panel Speaker at IEEE VTC-Spring 2023, EuCNC and 6G Summit 2023, IEEE PIMRC 2023 Workshop, and IEEE BlackSeaCom 2023, and Tutorial/Invited Speaker at IEEE Globecom 2023, IEEE WCNC 2023, EuCNC and 6G Summit 2023, and several other conferences and events globally. He has been involved in Organising Committees and chairing technical program such as at IEEE ICC 2024, IEEE WCNC 2023-24, IEEE WF-PST 2024, IEEE ICMLCN 2024, and many workshops such as at IEEE ICC 2024, IEEE Globecom 2023, IEEE WCNC 2023, IEEE PIMRC 2022-23, etc. https://sites.google.com /view/aryankaushik/
Prof. Carlo Fischione is Professor of Electrical Engineering and Computer Science at KTH, Sweden. He is the director of the KTH-Ericsson Data Science Degree Program and Chair of the IEEE Machine Learning for Communications ETI. He received Laurea (summa cum laude) in electronic engineering and Ph.D. degree in electrical and information engineering from the University of L’Aquila, Italy, in 2001 and 2005, respectively. He has held research positions at MIT, Cambridge, MA, USA, (2015 as a Visiting Professor); Harvard University, Cambridge, MA, USA, (2015 as an Associate Professor); and the University of California at Berkeley, CA, USA, (2004–2005 as a Visiting Scholar and 2007–2008 as a Research Associate). He received “IEEE ComSoc S. O. Rice” Award for the Best IEEE TCOM Paper of 2018, the Best Paper Award of IEEE TII in 2007, the Best Paper Awards at the IEEE MASS 2005 and 2009, the Best Paper Award of the IEEE Sweden VT-COM-IT Chapter in 2014, the Best Business Idea Awards from the VentureCup East, Sweden, in 2010, and the Stockholm Innovation and Growth (STING) Life Science in Sweden in 2014. He is the Co-Founder and the Scientific Director of ELK. https://people.kth.se/~carlofi/