Multi-channel free space optical convolutions

Abstract

Convolutional layers are a critical feature of modern neural networks and require significant com-putational resources. In response, optical accelerators have been developed as a low-energy, high-bandwidth approach for performing large-scale convolutions. We extend these methods to act on many input channels each with their own set of convolutional kernels. We simulate the performance of this system with ray-tracing and evaluate its performance.