Optical/Photonics Computing System (OPTSys) Seminar Series
A series of talks from experts in optics and photonics
In a world where Moore's law is encountering the physical limits and electronic circuits, after six decades of development, face formidable barriers, the pursuit of novel computing paradigms has become paramount. Among these frontiers, optical and photonic systems have emerged as a driving force. Yet, they present a unique challenge – the complexity of understanding optical/photonic systems, governed by Maxwell's equations, surpasses that of electronic circuits.
This is where our seminar series steps in! We have assembled some of the brightest minds from both industry and academia to bring the cutting-edge field of optical/photonic systems to your doorstep. Whether you are an enthusiastic junior student seeking to explore the wonders of this domain or a seasoned faculty member aiming to delve deeper, our seminars offer an ideal platform. Expect dynamic discussions, inspiration, and the potential for fruitful collaborations. Join us in shaping the future of computing. The future has arrived, and it's illuminated by optics!
To stay updated as an audience member, please subscribe to our mailing list. If you're interested in presenting, please reach out to us at optsys23@gmail.com. Explore our archive of past seminars, complete with recorded videos and presentation slides, to see the history of our engaging sessions.
Upcoming Seminar
Title: A Computer Engineering Journey to Optical Neural Networks: Infrastructure, Algorithms, and Co-design
Abstract: Despite the significant progress in customized ML/AI accelerator designs, the Pareto-frontier encompassing performance, energy efficiency, and carbon emissions of digital accelerators remains unchanged due to the reliance on conventional technologies. As an alternative, optical neural networks (ONNs), such as diffractive optical neural networks (DONNs), promise vast improvements in terms of computing speed, power efficiency, and carbon dioxide emissions. Nonetheless, designing and deploying DONNs face critical challenges. These primarily stem from the requirements for domain-specific infrastructure, algorithms, and the hurdles posed by multi-disciplinary domain knowledge in optical physics, fabrication, ML, and co-design. In this presentation, I will share our journey towards the automatic and agile design of DONNs. We address the challenges of building tangible DONN systems through multi-disciplinary developments encompassing physics, algorithms, co-design, and hands-on prototyping. I will begin by introducing the core concepts of DONNs and the associated design difficulties. Subsequently, I will detail our comprehensive design infrastructure, LightRidge, and the physics-aware hardware-software co-design algorithms that facilitate immediate DONN fabrication and deployment using physical prototypes. I will conclude with recent case studies showcasing the application in complex tasks, like autonomous driving, facilitated by ML-assisted architecture exploration.
Biography: Cunxi Yu is an Assistant Professor at the University of Maryland, College Park. His research interests focus on novel algorithms, systems, and hardware designs for computing and security. Before joining University of Maryland, Cunxi was Assistant Professor with University of Utah, and held PostDoc at Cornell University. His work received the Best Paper Award at DAC (2023), Best Paper Nominations at ASP-DAC (2017) and TCAD (2018), NSF CAREER Award (2021), and American Physical Society DLS poster award (2022). Cunxi earned his Ph.D. from UMass Amherst in 2017.
Date and Time: April 11, 2024, 10:30 am - 11:30 am (U.S. Eastern Time)
Zoom Link: https://mit.zoom.us/j/97117541600
Organizers
Acknowledgment
This seminar was made possible through the generous contributions of our presenters and the active participation of our audience. Particualrly, we would like to extend our gratitude to the following individuals for their invaluable help in refining this seminar and making it accessible to a wider audience: Hanrui Wang.