The convergence of artificial intelligence and programmable optics is opening new opportunities for both biological imaging and optical machine learning. By shifting complexity from optical hardware to data-driven algorithms and reconfigurable systems, these approaches promise faster, more adaptive, and more efficient information processing.
In this talk, I present two complementary directions in this emerging field. First, I introduce an AI-enhanced microscopy framework that reconstructs depth-resolved volumetric information from limited two-dimensional measurements. By learning strong three-dimensional priors with generative models, this method improves reconstruction quality while reducing acquisition time, enabling faster imaging of dynamic biological processes. Second, I discuss programmable optical processors for machine learning, highlighting reconfigurable optical architectures that enable nonlinear encoding and programmable optical matrix–vector multiplication using complex media.
Together, these results point toward scalable optical systems that can potentially transform high-speed bioimaging and energy-efficient machine learning.
Fei Xia is an Assistant Professor of Electrical Engineering and Computer Science (EECS), an affilated faculty member at Beckaman Laser Institute at the University of California, Irvine. She completed her Ph.D. at Cornell and her postdoc at École Normale Supérieure (ENS) in Paris. She leads a research team working at the interface of optics, computation, and the brain, with a particular interest in designing smart optical systems to image, sense, and process biomedical information probed by light. She aims to provide innovative solutions to challenging biomedical needs through the co-design of hardware and software. Fei has been recognized by several major awards and honors, including the NIH Venture Program Awardee, Scialog Fellow, Rising Star in Light, Optica Foundation Challenge Award, Seal of Excellence for Marie Skłodowska-Curie Fellowship, the SPIE Women in Optics, Data Open Championship and the Mong NeuroTech Fellowship.