Photonic Tensor Accelerators for Artificial Neural Networks

用于人工神经网络的光子张量加速器

• 批准号: 1932858
• 负责人: Guifang Li
• 金额: $ 47.5万
• 依托单位: The University of Central Florida Board of Trustees
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1932858&HistoricalAwards=false
• 关键词: Photonic; Tensor; Accelerators; Artificial; Neural

Artificial intelligence (AI) and artificial neural networks (ANNs) have dominated conversations about the future of science, technology, economy, society and culture, and even humanity itself. Excitement about AI comes about because it recognizes patterns and even discovers solutions that are superior to those based on human intelligence. The rapid progress of AI is greatly attributed to increased computing capabilities. In recent years, the computation power of integrated circuits (ICs) have been unable to sustain the growth rate according to Moore's law. Hardware accelerators, with processing units and optimized memory architecture designed specifically for parallel computing, played a key role in the implementation of machine learning (ML) models. However, electronic hardware accelerators have already been pushed to their limits in term of scalability, unable to keep up with the exponential growth of data volume. Against this backdrop, there have been renewed efforts in exploring the role of optics in computing, motivated by the large bandwidth and low loss of optical transmission. This project proposes the photonic tensor accelerator (PTA), a highly-parallel photonic architecture capable of matrix-vector multiplication and matrix-matrix multiplication, that offers a computing power several orders of magnitude higher than existing electronic accelerators. Thanks to its high degree of parallelization, PTA is specifically suited for batch matrix multiplication for the implementation of ANN models. The technology developed in this proposal could demonstrate the cooperative roles of advanced hardware and software and attract more students into hardware-related areas. The research proposed is interdisciplinary in nature and can serve as a platform for training both graduate and undergraduate students at UCF, a Hispanic Serving Institution (HSI) designated by the U.S. Department of Education.The overarching goal of this project is to construct photonic accelerators that 1) offer orders-of-magnitude higher scalability over electronics, 2) are fast, programmable, ideally compatible with training as well as inference, and 3) lower the power-consumption density to enable ANNs that are competitive over their pure electronic counterparts. The core of ANNs is tensor multiplication, which only require special operations (multiplication and accumulation, rather than general-purpose computing) in large scale that are especially suited for photonic accelerators. In addition, ANNs are robust to low dynamic range variabilities in nonlinear activation. PTA exploits all degrees of freedom of light to accelerate tensor multiplication. Specifically, PTA utilizes coherent beating between a signal and local oscillator to perform multiplication, frequency/wavelength, spatial modes and polarization for accumulation, and 2-D and 3-D parallelism of free space to scale the processing power. The proposed approach could scale the number of multiply-accumulate (MAC) operations by several orders of magnitude over the state-of-the-art IC hardware accelerators, including graphical processing units (GPUs) and ASICs such as tensor processing units (TPUs). The project will repurpose the technique of recirculating loops to scale up the number of layers for deep neural networks (DNNs). The proposed research is also synergistic with artificial intelligence (AI) in that some of the new devices will be designed using machine-learning techniques and the availability of the PTA-based ANNs allows new paradigms of ANN training.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

人工智能（AI）和人工神经网络（ANN）主导了有关科学、技术、经济、社会和文化甚至人类本身未来的对话。人工智能之所以令人兴奋，是因为它能够识别模式，甚至发现优于基于人类智能的解决方案。人工智能的快速进步很大程度上归功于计算能力的增强。近年来，集成电路（IC）的计算能力已经无法维持摩尔定律的增长速度。硬件加速器具有专为并行计算设计的处理单元和优化的内存架构，在机器学习 (ML) 模型的实施中发挥了关键作用。然而，电子硬件加速器在可扩展性方面已经达到极限，无法跟上数据量的指数增长。在此背景下，由于光传输的大带宽和低损耗，人们重新努力探索光学在计算中的作用。该项目提出了光子张量加速器（PTA），这是一种能够进行矩阵-矢量乘法和矩阵-矩阵乘法的高度并行光子架构，其计算能力比现有电子加速器高出几个数量级。由于其高度并行化，PTA 特别适合用于实现 ANN 模型的批量矩阵乘法。该提案中开发的技术可以展示先进硬件和软件的协作作用，并吸引更多学生进入硬件相关领域。拟议的研究本质上是跨学科的，可以作为培训 UCF 研究生和本科生的平台，UCF 是美国教育部指定的西班牙裔服务机构 (HSI)。该项目的总体目标是构建光子加速器，1) 提供比电子产品高几个数量级的可扩展性，2) 快速、可编程，与训练和推理完美兼容，3) 降低功耗密度，使人工神经网络比纯电子同类产品更具竞争力。人工神经网络的核心是张量乘法，只需要大规模的特殊运算（乘法和累加，而不是通用计算），特别适合光子加速器。此外，人工神经网络对于非线性激活中的低动态范围变化具有鲁棒性。 PTA 利用光的所有自由度来加速张量乘法。具体来说，PTA 利用信号和本地振荡器之间的相干拍频来执行乘法、频率/波长、空间模式和偏振以进行累积，以及自由空间的 2-D 和 3-D 并行性来扩展处理能力。所提出的方法可以将乘法累加 (MAC) 运算的数量比最先进的 IC 硬件加速器（包括图形处理单元 (GPU) 和张量处理单元 (TPU) 等 ASIC）扩展几个数量级。该项目将重新利用循环循环技术来扩大深度神经网络（DNN）的层数。拟议的研究还与人工智能 (AI) 具有协同作用，因为一些新设备将使用机器学习技术进行设计，并且基于 PTA 的 ANN 的可用性允许新的 ANN 训练范式。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力优点和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Vector-mode Multiplexing For Photonic Tensor Accelerator

光子张量加速器的矢量模式复用

• 发表时间: 2022
• 期刊: OptoElectronics and Communications Conference (OECC
• 作者: Alireza Fardoost, Fatemeh Ghaedi

Multiplane light conversion design with physical neural network

基于物理神经网络的多平面光转换设计

• 发表时间: 2022
• 期刊: Digital Holography and Three-Dimensional Imaging
• 作者: Zheyuan Zhu, Joe H.

Fabry-Perot Filter-Based Mode-Group Demultiplexers

基于法布里-珀罗滤波器的模式组解复用器

• DOI: 10.1364/cleo_si.2020.sth1l.3
• 发表时间: 2020
• 期刊: 2020 Conference on Lasers and Electro-Optics (CLEO
• 作者: Vanani, Fatemeh Ghaedi;Fardoost, Alireza;Li, Guifang
• 通讯作者: Li, Guifang

A Reconfigurable Broadband Space-Mode Router using Multiplane Light Conversion

使用多平面光转换的可重构宽带空间模式路由器

• DOI: 10.1109/ipc47351.2020.9252257
• 发表时间: 2020
• 期刊: 2020 IEEE Photonics Conference (IPC
• 作者: Zhang, Yuanhang;Wen, He;Fontaine, Nicolas K.;Chen, Haoshuo;LiKamWa, Patrick L.;Li, Guifang
• 通讯作者: Li, Guifang

Mode-Group Demultiplexers Using Thin-Film Filters

• DOI: 10.1109/ipc47351.2020.9252489
• 发表时间: 2020-09
• 期刊: 2020 IEEE Photonics Conference (IPC)
• 作者: Fatemeh Ghaedi;Vanani Alireza;Fardoost Guifang Li