Matrix computation, as a fundamental building block of information processing in science
and technology, contributes most of the computational overheads in modern signal …

Integrated photonic neural networks provide a promising platform for energy-efficient, high-
throughput machine learning with extensive scientific and commercial applications. Photonic …

Conventional electronic processors, which are the mainstream and almost invincible
hardware for computation, are approaching their limits in both computational power and …

We introduce an electro-optic hardware platform for nonlinear activation functions in optical
neural networks. The optical-to-optical nonlinearity operates by converting a small portion of …

As deep neural networks (DNNs) revolutionize machine learning, energy consumption and
throughput are emerging as fundamental limitations of CMOS electronics. This has …

Programmable photonic circuits of reconfigurable interferometers can be used to implement
arbitrary operations on optical modes, providing a flexible platform for accelerating tasks in …

For the benefit of designing scalable, fault resistant optical neural networks (ONNs), we
investigate the effects architectural designs have on the ONNs' robustness to imprecise …

Probabilistic machine learning utilizes controllable sources of randomness to encode
uncertainty and enable statistical modeling. Harnessing the pure randomness of quantum …

We experimentally demonstrate an on-chip electro-optic circuit for realizing arbitrary
nonlinear activation functions for optical neural networks (ONNs). The circuit operates by …

Unitary optical processors (UOPs) are task-specific computing units that can ultimately
enable ultrafast and energy-efficient unitary matrix-vector multiplications based on the …