Comsoc Distinguished Lecture:Machine Learning over Networks: From Fundamentals to Applications

#joint #chapter #edge #computing #AI #machine #learning
Share

IEEE Comsoc San Diego chapter Distinguished Lecture: 

     co-hosted by Comsoc Twin chapter Program (TwCP), San Diego and Bangalore Chapters 

 

Talk 1, in person,

Machine Learning over Networks: From Fundamentals to Applications

by Dr. Tamer ElBatt,  The American University in Cairo (AUC), Egypt

 

Talk 2,  pre-recorded (speaker is from India, Bangalore Chapter)

The Thinking Network: How AI is Rewiring the RAN

Dr. Madhan Raj Kanagarathinam, Samsung R&D Institute India - Bangalore

Here is the link to recorded talk, you can view here any time.

https://drive.google.com/drive/folders/1TpodLdwXZnFVAmsSc-Qt9EqcUWoLrzir

The link is accessible from any IEEE ID.



  Date and Time

  Location

  Hosts

  Registration



  • Add_To_Calendar_icon Add Event to Calendar
  • 6455 Lusk Blvd
  • San Diego, California
  • United States 92121
  • Building: Qualcomm Buidling Q
  • Room Number: Auditorium

  • Contact Event Hosts
  • Co-sponsored by Bangalore Chpater, San Diego Section, San Diego VTS chapter, CS chapter
  • Starts 15 August 2026 07:00 AM UTC
  • Ends 31 August 2026 11:00 PM UTC
  • No Admission Charge


  Speakers

Dr. Tamer ElBatt

Topic:

Machine Learning over Networks: From Fundamentals to Applications

In this talk, we review sample results based on our recent work in the areas of machine learning over networks and edge intellligence. In particular, we touch upon our work on communication-efficient federated learning and its applications. 

In the first part of the talk, we shed light on topics pertaining to federated learning. First, we overview a second-order federated learning algorithm, coined Fed-Sophia, with communication and computation efficiency merits. Next, we introduce our recent work on second order state synchronization (SOSS) to address the limitations of Fed-Sophia with non-IID data. Afterwards we overview MIRA which proposes a novel approach of federated multi-task learning for fine-tuning LLMs. 

In the second part of the talk, we shift focus to edge intelligence applications in digital health with particular focus on CVD prediction. Finally, we present our work on architecting, designing and demonstrating a system prototype for multi-tier intelligence for IoT.

Biography:

Tamer ElBatt is a Professor at the Computer Science and Engineering (CSE) Dept., The American University in Cairo (AUC), Egypt. He received the B.S. and M.S. degrees in EECE from Cairo University, Egypt, respectively, and the Ph.D. degree in ECE from the University of Maryland, College Park, USA in 2000. From 2000 to 2009 he was with major U.S. industry R&D, e.g., HRL Labs, Malibu, USA and Lockheed Martin ATC, Palo Alto, USA, at various positions. From 2009 to 2017, he served at the EECE Dept., Cairo University, as an Assistant Professor and later as an Associate and Full Professor. He also held a joint appointment with Nile University, Egypt from 2009 to 2017 and served as the Director of the Wireless Intelligent Networks Center (WINC) from 2012 to 2017. In July 2017, he joined the CSE Dept. at the American University in Cairo as an Associate Professor, where he is currently a Professor. He served as the CSE Director of the Graduate Program and later as the Associate Chair, from 2019 to 2022. He published more than 165 papers in major journals and international conferences. Dr. ElBatt holds seven issued U.S. patents and one WIPO patent.

Dr. ElBatt served on the TPC of numerous IEEE and ACM conferences. He served as an Executive TPC Vice Chair of the 6th ICCSPA’24, TPC Co-Chair of IEEE MELECON’26 and 5th ICCSPA’22, Tutorial Co-Chair of IEEE MECOM’25, Track Co-Chair of ITC-Egypt’22, Publicity co-chair of IEEE LCN’23 and VTC-Spring 2020, Demo Co-Chair of ACM Mobicom 2013 and the Publication Co-Chair of IEEE Globecom’12 and Mobiquitous’14. He is currently serving on the Editorial Board of Frontiers in Communications and Networks – Data Science for Communications and IEEE TCCN and served on the Editorial Board of IEEE TCCN, TMC and Wiley IJSC&N. He served on NSF and Fulbright review panels. Dr. ElBatt was a Visiting Professor at Politecnico di Torino in Aug. 2010, Sabanci University in Aug. 2013 and University of Padova in Aug. 2015. Dr. ElBatt is a ComSoc Distinguished Lecturer, the recipient of the Google Faculty Research Award in 2011, 2012 Cairo University Incentive Award in Engineering, 2015 State Incentive Award in Engineering and the 2025 AUC Faculty Merit Award for Excellence in Research. His research interests lie in the broad areas of modeling, performance analysis, design and optimization of wireless networks, mobile computing and IoT systems. He served as the IEEE Egypt Section Vice Chairman (2020-2023). Dr. ElBatt is a Fellow of the African Academy of Sciences (AAS) and a Senior Member of the IEEE.

 

Dr. Madhan Raj Kanagarathinam

Topic:

The Thinking Network: How AI is Rewiring the RAN

​The modern Radio Access Network (RAN) is undergoing a fundamental transformation: shifting from a passive transport pipe into an intelligent, service-aware substrate. This talk explores this dual evolution through two distinct paradigms: AI in RAN and RAN for AI (AI-Aware RAN).
​First, we address AI in RAN—leveraging embedded machine learning across the Near-RT and Non-RT RIC (RAN Intelligent Controller) to automate radio resource management, optimize predictive beamforming, and dynamically control energy-saving routines in Open RAN architectures.
​Second, we dive into RAN for AI through the lens of AI-Aware RAN. As Generative AI, agentic workloads, and multimodal streams proliferate, traditional Quality of Service (QoS) frameworks based solely on bandwidth and raw packet delay fail. AI-Aware RAN introduces mechanisms to detect AI traffic flows (prompts, completions, and token streams) at the ingress layer and prioritize token delivery over the air interface. By shifting network optimization toward AI-native metrics—such as Time-to-First-Token (TTFT) and Inter-Token Latency (ITL)—the network dynamically allocates Data Radio Bearers (DRBs) and MAC grants to ensure real-time agentic interactions without stalling token generation.

 

Biography:

 

Dr. Madhan Raj Kanagarathinam is a Software Architect and AI x Wireless Researcher at Samsung R&D Institute India - Bangalore, where he leads innovations in network protocols, QoS/QoE optimization, and AI-driven connectivity for smartphones. With over a decade of experience, he pioneered the world's first 1 Gbps mobile throughput using MPTCP (commercialized in Galaxy S6), AI-powered Intelligent Wi-Fi for real-time app enhancement (up to 110% video bitrate gains, 6x gaming latency reduction), eBPF-based on-device traffic classification, and Enhanced Mobile Hotspot features deployed in Galaxy S21 to S25 series.

A prolific innovator, Dr. Madhan holds 145+ patent filings (52 granted) on MPTCP, QUIC, L4S, Wi-Fi 7/8 QoS, and multipath transport, with 30+ IEEE publications including IEEE TNSM, IEEE IOTJ, IEEE TCE, IEEE Access, ICC, Globecom, and WCNC. He edited landmark books 6G Mobile Wireless Networks (70K+ accesses) and AI in Wireless for Beyond 5G Networks. His Ph.D. (IIT Madras, 2024; Br. C. Selvam Award winner) focused on "AI-Driven Traffic Management for Enhanced QoE in Smartphones," bridging academia-industry with real-world impact.

Dr. Madhan is a Senior Member, IEEE ComSoc Distinguished Lecturer, Vice Chair of IEEE YP R10 Bangalore, recipient of the IEEE ComSoc Best Young Professional (YP) Award in Industry, and winner of the IEEE GLOBECOM 2025 Best Paper Award. These recognitions celebrate his impactful contributions to practical networking innovations that enhance global smartphone connectivity and user experience.