Seminar Speaker: Yongmin Liu, Northeastern University

Accelerating the Design and Development of Engineered Photonic Materials Based on Deep Learning

Abstract

Over the past decades, we have witnessed tremendous progress and success in engineered photonic materials, including photonic crystals, plasmonic nanostructures, and metamaterials. For instance, by tailoring the geometry of the building blocks of metamaterials and engineering their spatial distribution, we can control the amplitude, polarization state, phase, and trajectory of light in an almost arbitrary manner [1,2]. However, the conventional physics- or rule-based approaches are insufficient for designing multifunctional and multidimensional metamaterials, because the degrees of freedom in the design space become extremely large. Deep learning, a subset of machine learning that learns multilevel abstraction of data using hierarchically structured layers, could substantially accelerate the development of complex metamaterials and other photonic structures with high efficiency, accuracy and fidelity [3]. In this talk, I will show that deep learning can help to unveil the highly nonintuitive and nonlinear structure-property relationships in metamaterials. After training, the developed deep learning models can simultaneously realize the forward prediction and inverse design of 2D and 3D metamaterials that exhibit pronounced chiral or anisotropic responses [4,5]. Using this data-driven approach, we demonstrate multifunctional meta-devices, such as flat lenses and holograms, which show up to eight controllable responses subjected to different combinations of working frequencies and polarization states [6]. More exciting advancements are expected when researchers with different backgrounds collectively contribute to this emerging field where photonics and artificial intelligence merge. References: [1] Y. M. Liu and X. Zhang, Chemical Society Review 40, 2494 (2011). [2] B. Xiong et al., Science 379, 294 (2023). [3] W. Ma et al., Nature Photonics, 15, 77 (2021). [4] W. Ma et al., ACS Nano 12, 6326 (2018). [5] W. Ma et al., Advanced Materials 31, 1901111 (2019). [6] W. Ma et al. Advanced Materials 34, 2110022 (2022).

Biography

Dr. Yongmin Liu obtained his Ph.D. from the University of California, Berkeley in 2009. He joined the faculty of Northeastern University in Boston in the fall of 2012, and currently, he is a full professor in the Department of Mechanical & Industrial Engineering and the Department of Electrical & Computer Engineering. Dr. Liu’s research interests include nano optics, nanoscale materials and engineering, plasmonics, metamaterials, biophotonics, and artificial intelligence. He has authored and co-authored more than 120 journal papers, including Science, Nature, Nature Photonics, Nature Nanotechnology, Advanced Materials, Physical Review Letters, and Nano Letters. Dr. Liu was a recipient of the Søren Buus Outstanding Research Award at Northeastern University (2024), the Faculty Fellow of College of Engineering at Northeastern University (2019), NSF CAREER Award (2017), Office of Naval Research Young Investigator Award (2016), SPIE DCS Rising Researcher Award (2016), 3M Non-Tenured Faculty Award (2016), and Chinese Government Award for Outstanding Students Abroad (2009). He has served as an editorial board member for Nano Convergence, PhotoniX, EPJ Applied Metamaterials, and Scientific Reports. He is a fellow of Optica (formerly OSA) and SPIE.