Integrated multi‐operand optical neurons for scalable and hardware‐efficient deep learning
Chenghao Feng, Jiaqi Gu, Hanqing Zhu, Shupeng Ning, Rongxing Tang, May H. Hlaing, Jason Midkiff, Sourabh Jain, David Z. Pan, Ray T. Chen
Abstract
Optical neural networks (ONNs) are promising hardware platforms for next-generation neuromorphic computing due to their high parallelism, low latency, and low energy consumption. However, previous integrated photonic tensor cores (PTCs) consume numerous single-operand optical modulators for signal and weight encoding, leading to large area costs and high propagation loss to implement large tensor operations. This work proposes a scalable and efficient optical dot-product engine based on customized multi-operand photonic devices, namely multi-operand optical neuron (MOON). We experimentally demonstrate the utility of a MOON using a multi-operand-Mach-Zehnder-interferometer (MOMZI) in image recognition tasks. Specifically, our MOMZI-based ONN achieves a measured accuracy of 85.89 % in the street view house number (SVHN) recognition dataset with 4-bit voltage control precision. Furthermore, our performance analysis reveals that a 128 × 128 MOMZI-based PTCs outperform their counterparts based on single-operand MZIs by one to two order-of-magnitudes in propagation loss, optical delay, and total device footprint, with comparable matrix expressivity.