使用集成光子张量核心进行并行卷积处理
J Feldmann1, N Youngblood2,3, M Karpov4
1Institute of Physics, University of Münster, Münster, Germany.
Nature
|January 7, 2021
概括
研究人员开发了一种光子张量核, 一种光学硬件加速器, 这种集成的光子设备为数据密集型应用提供了更快,更可扩展的AI硬件.
科学领域:
- 综合光学
- 光学计算
- 人工智能硬件
背景情况:
- 移动网络,物联网和人工智能的指数增长需要更快,更高效的硬件.
- 对于处理大量数据集的速度和可扩展性的现有硬件限制.
- 需要专门的硬件加速器来处理计算密集的AI任务.
研究的目的:
- 展示一个特定的集成光子硬件加速器 (电阻芯).
- 使用光子技术实现高速并行内存计算.
- 探索未来人工智能硬件中集成光子学的潜力.
主要方法:
- 开发了一种利用相变材料记忆阵列的光子张量核.
- 使用基于光子芯片的光频 (soliton微) 进行计算.
- 通过可重新配置的被动元件来测量光学传输.
主要成果:
- 实现了每秒数万亿个乘积运算的运行速度 (Tera-MACs/s).
- 已证明的计算带宽超过14千兆赫,受调节器和光探测器速度的限制.
- 展示了光子张量核心的CMOS晶片规模集成的途径.
结论:
- 在光学计算硬件的显著进步.
- 集成光学为平行,快速和高效的AI计算提供了一个有前途的解决方案.
- 这项技术在自动驾驶,实时视频处理和云计算方面具有潜在的应用.
更多相关视频
14:09High-Throughput Total Internal Reflection Fluorescence and Direct Stochastic Optical Reconstruction Microscopy Using a Photonic Chip
Published on: November 16, 2019
7.2K
10:21A Coregistered Ultrasound and Photoacoustic Imaging Protocol for the Transvaginal Imaging of Ovarian Lesions
Published on: March 3, 2023
2.0K
相关概念视频
Convolution Properties II
448
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
448
Convolution Properties I
379
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
379
Parallel Processing
444
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
444
Convolution: Math, Graphics, and Discrete Signals
673
In any LTI (Linear Time-Invariant) system, the convolution of two signals is denoted using a convolution operator, assuming all initial conditions are zero. The convolution integral can be divided into two parts: the zero-input or natural response and the zero-state or forced response, with t0 indicating the initial time.
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
673
Deconvolution
418
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
418
Computed Tomography
7.6K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
7.6K
