返回
Fiber optic computing using distributed feedback
DOI:10.1038/s42005-024-01549-1.png)
摘要
En 中文
The widespread adoption of machine learning and other matrix intensive computing algorithms has renewed interest in analog optical computing, which has the potential to perform large-scale matrix multiplications with superior energy scaling and lower latency than digital electronics. However, most optical techniques rely on spatial multiplexing, requiring a large number of modulators and detectors, and are typically restricted to performing a single kernel convolution operation per layer. Here, we introduce a fiber-optic computing architecture based on temporal multiplexing and distributed feedback that performs multiple convolutions on the input data in a single layer. Using Rayleigh backscattering in standard single mode fiber, we show that this technique can efficiently apply a series of random nonlinear projections to the input data, facilitating a variety of computing tasks. The approach enables efficient energy scaling with orders of magnitude lower power consumption than GPUs, while maintaining low latency and high data-throughput. Optical techniques adopted in optical computing rely on spatial multiplexing, requiring numerous integrated elements and restricting the architecture to perform a single kernel convolution per layer. The authors demonstrate a fiber-optic computing architecture based on temporal multiplexing that performs multiple convolutions in a single layer.
Keyword:
EXTREME LEARNING-MACHINE
ARTIFICIAL-INTELLIGENCE
RECOGNITION
期刊
IF:
5.8
论文数:
2.9K
被引数:
1.1W
机构
引用论文
Analytical model and experimental verification of the critical power for modulation instability in optical fibers
OPTICS EXPRESS
IF3.3
Productivity enhancement of solar still by PCM and Nanoparticles miscellaneous basin absorbing materials
Desalination
IF0
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
Photonics for artificial intelligence and neuromorphic computing用于人工智能和神经形态计算的光子学
NATURE PHOTONICS
IF32.9
Femtofarad optoelectronic integration demonstrating energy-saving signal conversion and nonlinear functions
NATURE PHOTONICS
IF32.9


