Shadi Noghabi is a senior researcher in the Networking Research Group at Microsoft Research. She received her Ph.D. Degree in CS from the University of Illinois at Urbana-Champaign in 2018 where she worked on the design and development of several large-scale production systems. Specifically, in collaboration with LinkedIn, she developed Ambry, a geo-distributed storage system used as the main backed for media objects at LinkedIn, and Samza, an Apache open-source stream processing framework used by LinkedIn and 10s of other companies. She has received the David Kuck Outstanding Thesis Award and the Microsoft Research Dissertation Award for her work during her Ph.D. Her current focus at Microsoft is on edge computing and Industrial IoT. She is currently working on developing highly available and scalable management platforms for the edge, ranging from small edges in remote areas to large renewable-based edge data centers and edge computing in space.

Abstract: Edge computing, where compute capacity is placed close to the data generation, has become an increasingly popular architecture. Using an edge enables lower latency, reduced bandwidth consumption, continuous computation despite disconnection from the cloud, and data privacy, all of which have created the widespread penetration of edge computing across many industries ranging from enterprise scenarios, manufacturing, agriculture, to 5G networks. However, in many such scenarios, the edge is not equivalent to a cloud data center with large reliable compute and networking infrastructure. Often, edges are in remote locations, with intermittent and weak network connectivity alongside limited and unreliable compute capacity. We refer to such edge setups as the "extreme edge".

In this talk, we will cover three such extreme edge scenarios, each with its own unique constraints, and describe the techniques used to cope with such constraints. First, an edge data center scenario deployed right at renewable farms and powered solely by renewables where the compute capacity varies based on the power production. Second, an edge in the farm scenario with limited network connectivity to the cloud while requiring the processing and uplink of drone imagery in near real-time. Finally, the uprising area of edge in Space (LEO satellite constellations) with both intermittent network and compute while capturing orders of magnitude more data than the bandwidth available.

Yiran Chen received B.S. (1998) and M.S. (2001) from Tsinghua University and Ph.D. (2005) from Purdue University. After five years in industry, he joined University of Pittsburgh in 2010 as Assistant Professor and then was promoted to Associate Professor with tenure in 2014, holding Bicentennial Alumni Faculty Fellow. He is now the Professor of the Department of Electrical and Computer Engineering at Duke University and serving as the director of the NSF AI Institute for Edge Computing Leveraging the Next-generation Networks (Athena), the NSF Industry–University Cooperative Research Center (IUCRC) for Alternative Sustainable and Intelligent Computing (ASIC), and the co-director of Duke Center for Computational Evolutionary Intelligence (DCEI). His group focuses on the research of new memory and storage systems, machine learning and neuromorphic computing, and mobile computing systems. Dr. Chen has published 1 book and about 500 technical publications and has been granted 96 US patents. He has served as the associate editor of more than a dozen international academic periodicals and served on the technical and organization committees of more than 60 international conferences. He is now serving as the Editor-in-Chief of the IEEE Circuits and Systems Magazine. He received eight best paper awards, one best poster award, and fourteen best paper nominations from reputable international conferences and workshops such as MICRO, KDD, DATE, SEC, etc. He received numerous awards for his technical contributions and professional services such as IEEE Computer Society Edward J. McCluskey Technical Achievement Award, ACM SIGDA Service Award, etc. He is a Fellow of the ACM and IEEE and now serves as the chair of ACM SIGDA.

Abstract: Federated learning has become a popular distributed machine learning paradigm for developing on-device AI applications. However, the data residing across devices is intrinsically statistically heterogeneous (i.e., non-IID data distribution) and mobile devices usually have limited communication bandwidth to transfer local updates. Such statistical heterogeneity and communication limitation are two major bottlenecks that hinder applying federated learning in practice. In addition, recent works have demonstrated that sharing model updates makes federated learning vulnerable to inference attacks and model poisoning attacks. In this talk, we will present our recent works on novel federated learning frameworks to address the scalability and heterogeneity issues simultaneously. In addition, we will also reveal the essential reason of privacy leakage and model poisoning attacks in federated learning procedures, and provide the defense mechanisms accordingly towards trustworthy federated learning.