.png){kind=icon}
返回
Direct Optical Convolution Computing Based on Arrayed Waveguide Grating Router
.png){kind=icon}
PRE
AI
DOI：10.1002/lpor.202301221.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Optical convolution computing is gaining traction owing to its inherent parallelism, multi-dimensional processing, and energy efficiency. To handle input dimensions of N, conventional implementations necessitate N2 optical elements, such as Mach-Zehnder interferometers or micro-ring resonators, to process multiply-accumulate (MAC) operations, limiting scalability and resulting in elevated power consumption. Here, a direct convolution computing method based on wavelength routing, utilizing the unique sliding property of an arrayed waveguide grating router (AWGR) to perform the sliding window operation of the convolution in the wavelength-space domains is proposed. With two input vectors directly loaded onto two modulator arrays, the convolution result is instantaneously produced at a photodetector array. The entire convolution computation is executed within a single clock cycle without the need for preprocessing or decomposition into elementary MAC operations. The number of active elements is minimal, only needed for input/output. The proposed optical convolution unit has striking advantages of high scalability, high speed, and processing simplicity compared to those based on optical matrix-vector multipliers. In the first experimental demonstration, a remarkable classification accuracy of up to 98.2% in handwritten digit recognition tasks using a LeNet-5 neural network is achieved. A direct convolution computing method based on wavelength routing, utilizing the unique sliding property of an arrayed waveguide grating router (AWGR) to perform the sliding window operation of the convolution in the wavelength-space domains is proposed and demonstrated. With two inputs vectors directly loaded onto modulator arrays, the convolution result is instantaneously produced at a photodetector array. image
Keyword:
arrayed waveguide grating router
handwritten digit recognition
LeNet-5 neural network
optical convolution computing
期刊
L
IF:
10
论文数:
3.7K
被引数:
2.1W
机构
引用论文
The cathodic behavior of microwave-treated manganese dioxide in lithium nonaqueous cells微波处理的二氧化锰在非水锂电池中的阴极行为
PREAI
分享收藏Compact optical convolution processing unit based on multimode interference基于多模干涉的紧凑型光学卷积处理单元.png)
NATURE COMMUNICATIONS
OAAI
NATURE COMMUNICATIONSIF15.7
OAAI分享收藏Training of photonic neural networks through in situ backpropagation and gradient measurement通过原位反向传播和梯度测量训练光子神经网络OPTICAOAAI
OPTICAIF8.5
OAAI分享收藏
查看更多内容