Abstract

Abstract

Abstract

Machine learning (ML) algorithms automate the data-label mapping and decision-making processes. The output of the ML algorithms determines the next step in real-world applications. It is desirable to have ML-based systems that are simultaneously accurate, secure, energy-efficient, low-cost, silently operable, maintainable, customizable, low-delay, and scalable. However, there are tradeoffs among these design goals that impact overall system complexity. In this talk, we will be focusing on the accuracy, security, and energy-efficiency objectives.

Abstract

Nanofabrication has enabled rapid development of state-of-the-art electronic and optical devices. With nanofabricated feature size going down to become comparable, or even smaller than, the visible wavelength, some unique optical properties, which are different from their macroscale behaviours, appear. Nanostructure engineering enables us to exploit these unique properties to achieve highly efficient, compact and cheap optical devices.