Si microring resonator crossbar arrays for deep learning accelerator

Abstract

We propose Si microring resonator (MRR) crossbar arrays as a programmable nanophotonoic processor (PNP) for a deep learning accelerator. The proposed MRR crossbar array can perform multiply-accumulate (MAC) operation in an optical domain. We numerically reveal that an optical neural network (ONN) based on the proposed MRR crossbar arrays can be used for the pattern recognition task with a similar performance to that of an ONN based on cascaded Mach–Zehnder interferometers. We predict that the power consumption can be reduced by approximately tenfold. The small area of the MRR also contributes to the reduction in its chip size by a factor of 36. The fabricated test devices consisting of Si MRRs with phase shifters exhibited no significant crosstalk between neighboring MRRs and showed the feasibility of MAC operation. Photonic integrated circuit using the proposed MRR crossbar arrays is promising for large-scale and low-power PNP for deep learning.

Arbitrary linear transformations for photons in the frequency synthetic dimension

“…Our work is in contrast to previous works where different frequency channels were used in parallel but without frequency conversions among them [6,7,9,11] by demultiplexing the different frequencies into separate spatial channels. Additionally, optimized fast modulation has been used for tailoring single photon spectra from two-level quantum emitters [35], or for quantum frequency conversion [15] and linear optical quantum computation [17,36], where the modulator is used as a generalized beam splitter in synthetic frequency dimensions.…”

“…Arbitrary linear transformations in photonics [1][2][3] are of central importance for optical quantum computing [4], classical signal processing and deep learning [5][6][7][8][9][10]. A variety of architectures are being actively studied to implement linear transformations for quantum computation and photonic neural networks, including those based on Mach-Zender interferometers (MZI) [4,5], microring weight banks [6,7,11], phase-change materials [8,9], and diffractive metasurfaces [10]. All such approaches use path encoding of photons in real space.…”

Si Microring Resonator Crossbar Array for On-Chip Inference and Training of the Optical Neural Network

“…Furthermore, our chip allows the direct implementation of transpose matrices without reconfiguring the MRRs, thus can be used to accelerate both the training and the inference of deep learning. In contrast to our previous work, in which we preliminarily investigated the characteristics of a 3 × 3 MRR crossbar array, , the chip in this work fully integrates a 4 × 4 MRR crossbar array, two input modulator arrays (one for inference signals and one for training signals), and Ge photodetectors (PDs). We experimentally demonstrate a simple inference task of image recognition using this ONN chip, obtaining a recognition accuracy of 93% with a pretrained three-layer neural network.…”

Arbitrary linear transformations for photons in the frequency synthetic dimension

“…Arbitrary linear transformations in photonics 1 – 3 are of central importance for optical quantum computing 4 , classical signal processing and deep learning 5 – 10 . A variety of architectures are being actively studied to implement linear transformations for quantum computation and photonic neural networks, including those based on Mach–Zender interferometers (MZI) 4 , 5 , microring weight banks 6 , 7 , 11 , phase-change materials 8 , 9 , and diffractive metasurfaces 10 . All such approaches use path encoding of photons in real space.…”

“…Previous works have considered implementing photonic linear transformations using different frequency channels in parallel but without frequency conversions among them 6 , 7 , 9 , 11 by demultiplexing the different frequencies into separate spatial channels. Additionally, optimized fast modulation has been used for tailoring single photon spectra from two-level quantum emitters 35 , or for quantum frequency conversion 15 and linear optical quantum computation 17 , 36 , where the modulator is used as a generalized beam splitter in synthetic frequency dimensions.…”