Abstract: Effectively learning from the ever-expanding pool of data generated by IoT devices poses a challenge, mainly due to data regulations and privacy concerns. Federated learning shows promise as a method for collaboratively training models on edge devices without exposing sensitive data. However, deploying federated learning in real-world IoT networks remains challenging due to the heterogeneity of systems and data, as well as the coexistence of multiple jobs. This talk will present a systematic solution, HEAL, for Heterogeneity-aware Efficient and Adaptive Learning for concurrent jobs in a shared IoT network, from the two perspectives: 1) the adaptive offloading of training computation from heterogeneous edge devices that can strike a balance between computation, communication, and privacy leakage risk; 2) the judicious coordination of edge devices in the distributed training procedures of multiple concurrent learning jobs, aiming for system efficiency and model quality. This talk will also present case studies on smart healthcare applications and other trending areas to highlight practical implications and insights.
Bio: Dr. Li Chen (https://lichenut.github.io) is an Associate Professor at the School of Computing and Informatics from the University of Louisiana at Lafayette. She received her Ph.D. from the Department of Electrical and Computer Engineering at the University of Toronto in 2018. She is an NSF EPSCoR Research Fellow, and an IEEE Senior Member. Her research interests include cloud computing, distributed deep learning, and AI applications. Dr. Chen is currently a visiting scholar at UIUC, hosted by Prof. Klara Nahrstedt.