Title: Photonic Resonator Networks: From Complex Phenomena to Computing and AI
Abstract: Over the past few decades, photonics has played a crucial role in applications ranging from long-haul ultrafast telecommunications to optical sensing and spectroscopy. The sheer volume of the data generated from the widespread implementation of such technologies is rapidly growing to levels where the conventional computing hardware architectures are failing to catch up. The resulting bottleneck strongly suggests the need for a revolution in the computing hardware that goes beyond the traditional von-Neumann CMOS architectures. In this regard, the emerging field of optical computing aims to effectively harness the high bandwidths offered by light for faster and more efficient application-specific information processing. In this talk, I will discuss my recent work on developing a new optical computing platform that encodes information using ultrafast optical pulses, each playing the role of a resonator within a time-multiplexed network. This time-multiplexed platform is highly programmable, can be readily expanded to very large-scale networks, and can operate at ultrafast timescales. First, I will present our results on using this architecture to experimentally study emergent complex phenomena that arise from the combined effects of non-trivial topology and dissipation, such as non-Abelian dynamics and topological phase transitions. As an application of these dissipative topological models, I will present a new enhanced sensing mechanism that I experimentally demonstrated by utilizing the scalability of our time-multiplexed platform. Next, I will talk about our recent results on utilizing the ultrafast dynamics within the time-multiplexed resonator network to implement optical neural networks that can operate with very low latencies. Finally, I will discuss future directions on how this platform can be implemented on photonic integrated circuits based on thin-film lithium niobate in order to develop application-specific photonic integrated circuits (ASPICs) that can solve various computing tasks such as large-scale optimization problems or serve as optical neural networks that operate faster and more efficiently than the state-of-the-art electronic processors.
About the Speaker: Midya Parto is a postdoctoral research associate at the California Institute of Technology. His research focuses on emergent phenomena in optics that arise from the synergy among non-Hermitian and topological phenomena and optical nonlinear processes, with applications in on-chip light sources and optical information processing and computing. He received his PhD in Optics and Photonics from CREOL in 2019, and his Master’s and Bachelor’s degrees in Electrical Engineering from University of Tehran in 2014 and 2011, respectively. He is the recipient of several awards and fellowships including the Incubic/Milton Chang Grant from Optica in 2019 and the Graduate Dean’s Fellowship from the University of Central Florida from 2014 to 2018, in addition to being the Tingye Li Innovation Prize finalist in 2018. He has served as a program committee member for Optica Nonlinear Optics (NLO) Meeting in 2023.
Read More