The need for light manipulation on the nanoscale has prompted the recent advent and prosperity of plasmonic metastructures. To fully unlock the potential of such engineered optical media, plasmonic structures are exploited with progressively greater complexity, including those with arbitrarily complicated topology, spatially variant building blocks, and multi-layered configurations. The astronomical degrees of freedom associated with such structures have obstructed effective design and implementation schemes based on the conventional wisdom. We have developed a series of deep-learning enabled generative frameworks for the inverse design of plasmonic structures in response to on-demand optical properties, with extended case studies and experimental demonstrations.
|