In order to maintain privacy-sensitive data and to facilitate collaborative machine learning (ML) among distributed nodes, Federated Learning (FL) has emerged as an attractive paradigm, where local nodes collaboratively train a task model under the orchestration of a central server without accessing end-user data. However, the non-independent and-identically-distributed (non-IID) data samples across participating nodes slow model training and impose additional communication rounds for FL to converge. In this talk, I will present our recent efforts in addressing this issue, aiming to accelerate model convergence under the presence of nodes with non-IID dataset. Firstly, we propose an adaptive weighting strategy that assigns weight proportional to node contribution instead of according to the size of local datasets. It can reinforce positive (suppress negative) node contribution dynamically, leading to a significant communication round reduction. Secondly, we design a probabilistic node selection scheme that can preferentially select nodes to boost model convergence of FL with non-i.i.d. datasets. The proposed scheme adjusts the probability for each node to be selected in each round based on measuring the relationship between the local gradient and the global gradient from participating nodes. The superiority of the proposed approaches over the commonly adopted Federated Averaging (FedAvg) algorithm has been verified by extensive experimental results. 

Biography:

Dr. Ping Wang is an Associate Professor at the Department of Electrical Engineering and Computer Science, York University, and a Tier 2 York Research Chair. Prior to that, she worked with Nanyang Technological University, Singapore, from 2008 to 2018. Her research interests are mainly in the area of wireless communication networks, cloud computing and Internet of Things with the recent focus on integrating Artificial Intelligence (AI) techniques into communications networks. She has published more than 250 papers/conference proceedings papers. Her scholarly works have been widely disseminated through top-ranked IEEE journals/conferences and received the Best Paper Awards from IEEE Wireless Communications and Networking Conference (WCNC) in 2022, 2020 and 2012, from IEEE Communication Society: Green Communications & Computing Technical Committee in 2018, and from IEEE International Conference on Communications (ICC) in 2007. Her work received 21,000+ citations with H-index 70 (Google Scholar). She is an IEEE Fellow and a Distinguished Lecturer of the IEEE Vehicular Technology Society.

Vehicle-to-Infrastructure (V2I) communication is becoming critical for the enhanced reliability of autonomous vehicles (AVs). However, the uncertainties in the road-traffic and AVs' wireless connections can severely impair timely decision-making. It is thus critical to simultaneously optimize the AVs' network selection  and  driving policies in order to minimize road collisions while maximizing the communication data rates. This talk will demonstrate  a reinforcement learning (RL) framework to characterize efficient network selection  and autonomous driving policies in a multi-band vehicular network (VNet) operating on conventional sub-6GHz spectrum and Terahertz (THz) frequencies.  The proposed framework is designed to maximize the traffic flow and minimize collisions by controlling the vehicle's motion dynamics (i.e., speed and acceleration) from autonomous driving perspective, and  maximize the data rates and minimize handoffs  by jointly controlling the vehicle's motion dynamics and network selection from telecommunication perspective.  Numerical results demonstrate interesting insights related to the inter-dependency of vehicle's motion dynamics, handoffs, and the communication data rate. The proposed policies enable AVs to  adopt safe driving behaviors with improved connectivity.

Biography:

Hina Tabassum is an Assistant Professor at the Lassonde School of Engineering, York University, Canada. Prior to that, she was a PDF at the Department of ECE, University of Manitoba, Canada. She received her PhD degree from King Abdullah University of Science and Technology (KAUST) in 2013. She is a Senior member of IEEE and a P.ENG in the province of Ontario. She has published over 70 technical articles in well-reputed IEEE journals and conferences. She is the founding chair of a special interest group on THz communications in IEEE ComSoc: Radio Communications Committee (RCC). She has been recognized as an Exemplary  Editor by IEEE Communications Letters, 2020, and an Exemplary Reviewer (Top 2% of all reviewers) by IEEE Transactions on Communications in 2015-2017, 2019, and 2020. Currently, she is serving as an Associate Editor in IEEE Communications Letters, IEEE Transactions on Green Communications, IEEE Communications Surveys and Tutorials, and IEEE Open Journal of Communications Society.  Her research interests include stochastic modeling, analysis, and optimization of energy efficient multi-band 5G/6G wireless networks jointly operating on sub-6GHz, millimeter, and Terahertz frequencies with applications to vehicular, aerial, and satellite networks.