Small microring array enables large complex-valued matrix multiplication

Optical computing uses photons instead of electrons to perform computations, which can significantly increase the speed and energy efficiency of computations by overcoming the inherent limitations of electrons. The basic principle of optical computing is the light-matter interaction. Matrix computing has become one of the most widely used and indispensable information processing tools in science and engineering, contributing a large number of computational tasks to most signal processing, such as discrete Fourier transforms and convolution operations. As the basic building block of artificial neural networks (ANNs), matrix multiplication occupies most of the computational resources. Due to the properties of electronic components, performing simple matrix multiplications require a large number of transistors to work together, while matrix multiplications can be easily implemented by fundamental photonic components such as microring, Mach Zehnder interferometer (MZI), and diffractive plane. Therefore, the speed of optical computing is several orders of magnitude faster than electronic computing and consumes much less power. However, the traditional incoherent matrix-vector multiplication method focuses on real-valued operations and does not work well in complex-valued neural networks and discrete Fourier transforms.

Researchers led by Prof. Jianji Dong at Huazhong University of Science and Technology (HUST), China, have proposed a photonic complex matrix-vector multiplier chip that can support arbitrary large-scale and complex-valued matrix multiplications. The chip breaks the bottleneck that traditional non-coherent optical computing schemes are difficult to achieve arbitrary large-scale complex-valued matrix multiplications, and also enables artificial intelligence applications such as discrete Fourier transform, discrete cosine transform, Walsh transform and image convolution. Their idea is to design matrix decomposition and matrix partitioning intelligent algorithms for the microring array architecture to extend matrix multiplications from real to complex domain and from small scale to large scale. The researchers successfully experimentally demonstrated several typical artificial intelligence applications, showing the great potential of the photonic complex matrix-vector multiplier chip for applications in artificial intelligence computing. The work entitled “A small microring array that performs large complex-valued matrix-vector multiplication” was published on Apr. 28 2022 in Frontiers of Optoelectronics.

Reference: Junwei Cheng, Yuhe Zhao, Wenkai Zhang, Hailong Zhou, Dongmei Huang, Qing Zhu, Yuhao Guo, Bo Xu, Jianji Dong, Xinliang Zhang. A small microring array that performs large complex-valued matrix-vector multiplication. Front. Optoelectron. 15, 15 (2022). https://doi.org/10.1007/s12200-022-00009-4

About Higher Education Press

Founded in May 1954, Higher Education Press Limited Company (HEP), affiliated with the Ministry of Education, is one of the earliest institutions committed to educational publishing after the establishment of P. R. China in 1949. After striving for six decades, HEP has developed into a major comprehensive publisher, with products in various forms and at different levels. Both for import and export, HEP has been striving to fill in the gap of domestic and foreign markets and meet the demand of global customers by collaborating with more than 200 partners throughout the world and selling products and services in 32 languages globally. Now, HEP ranks among China’s top publishers in terms of copyright export volume and the world’s top 50 largest publishing enterprises in terms of comprehensive strength.

The Frontiers Journals series published by HEP includes 28 English academic journals, covering the largest academic fields in China at present. Among the series, 13 have been indexed by SCI, 6 by EI, 2 by MEDLINE, 1 by A&HCI. HEP’s academic monographs have won about 300 different kinds of publishing funds and awards both at home and abroad.

About Frontiers of Optoelectronics

Frontiers of Optoelectronics (FOE) aims at introducing the most recent research results and the cutting edge improvements in the area of photonics and optoelectronics. It is dedicated to be an important information platform for rapid communication and exchange between researchers in the related areas. The journal publishes review articles, research articles, letters, comments, special issues, and so on. The Editors-in-Chief are Academician Qihuang Gong from Peking University and Prof. Xinliang Zhang from Huazhong University of Science and Technology. FOE has been indexed by ESCI, Ei, SCOPUS, CSCD, Source Journals for Chinese Scientific and Technical Papers and Citations, etc. FOE is fully open access since 2022.