Integrated multi-operand optical neurons for scalable and hardware-efficient deep learning
This work addresses hardware bottlenecks for scalable neuromorphic computing, offering a domain-specific improvement in optical deep learning systems.
The paper tackled the scalability and efficiency issues in optical neural networks by proposing multi-operand optical neurons (MOON), achieving 85.89% accuracy on the SVHN dataset with 4-bit precision and reducing propagation loss, delay, and footprint by orders of magnitude compared to single-operand methods.
The optical neural network (ONN) is a promising hardware platform for next-generation neuromorphic computing due to its 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 neurons (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 128x128 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.