Seminars
The Department of Electrical and Computer Engineering invites prestigious colleagues to speak during weekly graduate seminars. All talks take place from 12:00 pm–1:00 pm. Please see below for venue details.
For questions, please contact the committee chair, Tze Meng Low.
View all previously recorded seminars here. Andrew ID and password are required to view recorded seminars.
Sign up to receive notifications about ECE graduate seminars.
Graduate Seminars
All in-person seminars will follow CMU's gathering requirements in place at the time of the seminar. For those seminars taking place virtually, attendees will receive an email before the seminar with login information.
Box lunches/waters will be provided for the in-person seminars at 11:30 am in Panther Hollow Room, CIC 4th floor.
January 30 - Yasaman Ghasempour
Assistant Professor
Electrical and Computer Engineering
Princeton University
Title: Physics-guided Wireless Communication and Sensing Above 100 GHz
Abstract: The mmWave and sub-THz spectrum is rapidly emerging as a foundation for next-generation wireless communication and sensing systems, driven by its immense available bandwidth and sub-millimeter wavelengths. Yet, practical deployments face fundamental challenges: severe propagation loss, susceptibility to blockage, power-demanding PHY, and the breakdown of traditional far-field assumptions. Unlocking the full potential of these frontier frequencies demands physics-native solutions that capitalize on the unique properties of signals in these regimes. In this talk, I will first present an ultra-wideband retro-directive backscatter architecture above 100 GHz that departs from conventional large-scale antenna arrays and significantly reduces the power consumption. I will then discuss how the migration to higher frequencies, together with electronically large arrays, has extended the Fraunhofer limit from a few centimeters to several meters—placing many users into the electromagnetic near-field of future base stations and access points. Despite decades of progress in wireless communications, this near-field regime remains largely unexplored. I will show how programmable near-field beam shaping unlocks exciting new opportunities for communication and sensing. In particular, I will present AI-assisted self-curving beams that bend around obstacles in the environment, offering a path toward the long-standing vision of seamless connectivity in the presence of dynamic blockages, and provide tremendous potential for around-the-corner imaging. Finally, I will conclude by highlighting unprecedented application domains of mmWave/sub-THz sensing and imaging across disciplines such as agriculture and robotics, underscoring the transformative potential of these frontier bands.
Bio: Yasaman Ghasempour is an Assistant Professor of Electrical and Computer Engineering at Princeton University. She received her Ph.D. and master’s degree from Rice University and her bachelor’s degree from the Sharif University of Technology. Yasaman is the recipient of the Zhengyi Wang Prize (2026), Princeton Early-Career Faculty Award (2024), the AFOSR YIP Award (2024), the NSF CAREER Award (2022), the 2020 Marconi Young Scholar Award, and the Excellence in Teaching Award from Princeton School of Engineering and Applied Sciences. She has been named by the National Academy of Engineering (NAE) as one of the early-career Frontiers in Engineering. Yasaman is also listed as one of ten rising stars in communication and networking by N2Women. Her research received several Best Paper Awards, including USENIX NSDI, ACM MobiCom, ACM SenSys, and IEEE WCNC. Yasaman is the co-director of Princeton NextG Industry Affiliates Program. She serves on the TPC of several ACM SIGMOBILE conferences and is on the editorial board of Nature Communications Engineering, IEEE Transactions on Wireless Communications, IEEE Communications Magazine, and IEEE Journal of Infrared, Millimeter, and Terahertz Waves. Yasaman is featured in the Smithsonian Institutions Museum of Natural History as a change-making innovator in wireless technology. Her research is focused on next-generation wireless networks and sensing systems, including novel physical layer designs and link layer protocols for emerging wireless systems.
February 6 - Charith Mendis
Associate Professor
Siebel School of Computing and Data Science
University of Illinois at Urbana-Champaign
Title: Agile and Evolvable Software Construction in the Era of Rapidly Evolving Hardware Accelerator Designs
Abstract: Modern AI workloads have become exceedingly abundant and important in the current computing landscape. As a result, numerous software and hardware innovations have been developed to accelerate these workloads. However, we observe a subtle disconnect between the software and hardware communities. Most software innovations target well-established hardware platforms, such as CPUs (e.g., x86, ARM) and GPUs (e.g., NVIDIA GPUs). In contrast, hardware innovations produce numerous other tensor accelerator designs (e.g., Gemmini, Feather, Trainium) each year. We asked the question, why aren’t the software community using these accelerators or even evaluating them? The simple yet undeniable reason is the lack of standardized software tooling compared to CPUs and GPUs. For an architecture to be used, properly designed compiler backends, fast and scalable correctness and performance testing tools should be readily available (e.g., the CUDA ecosystem). In this talk, I will describe how we bridge this gap by automatically generating the necessary software tools for a large class of accelerators using the Accelerator Compiler Toolkit (ACT) ecosystem. Central to ACT is an ISA definition language, TAIDL, that for the first time standardizes how we define hardware-software interfaces for a large class of accelerators. Departing from the traditional approach of manually constructing test oracles, performance models, or retargetable compiler backends, we introduce agile, evolvable methodologies to automatically generate this necessary tooling using both formal methods and machine learning techniques for any TAIDL-defined accelerator interface. I will show how such automation enables rapid software prototyping, making rapidly evolving accelerator designs usable by the software community.
Bio: Charith Mendis is an Assistant Professor in the Siebel School of Computing and Data Science at the University of Illinois at Urbana-Champaign. His broad research interests are at the intersection of compilers, programming languages, and machine learning. He received his Ph.D. and Master’s from the Massachusetts Institute of Technology and his B.Sc. from the University of Moratuwa. He is the recipient of the DARPA Young Faculty Award, the NSF CAREER Award, the Google ML and Systems Junior Faculty Award, the Outstanding Advisor award at UIUC, the William A. Martin Outstanding Master’s Thesis Award at MIT, and the University Gold Medal for his B.Sc. He has won numerous paper awards, including a Distinguished Paper Award at POPL, a Best Student Paper Award at the IEEE BigData conference, an honorable mention for the Best Artifact Award at SIGMOD, a Best Paper Award at ML for Systems workshop at ISCA, and an IEEE Top Picks Honorable Mention.