Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.01117 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908295767261184 |
|---|---|
| author | Yang, Hangbo Peserico, Nicola Li, Shurui Ma, Xiaoxuan Schwartz, Russell L. T. Hosseini, Mostafa Babakhani, Aydin Wong, Chee Wei Gupta, Puneet Sorger, Volker J. |
| author_facet | Yang, Hangbo Peserico, Nicola Li, Shurui Ma, Xiaoxuan Schwartz, Russell L. T. Hosseini, Mostafa Babakhani, Aydin Wong, Chee Wei Gupta, Puneet Sorger, Volker J. |
| contents | Convolutional operations are computationally intensive in artificial intelligence services, and their overhead in electronic hardware limits machine learning scaling. Here, we introduce a photonic joint transform correlator (pJTC) using a near-energy-free on-chip Fourier transformation to accelerate convolution operations. The pJTC reduces computational complexity for both convolution and cross-correlation from O(N4) to O(N2), where N2 is the input data size. Demonstrating functional Fourier transforms and convolution, this pJTC achieves 98.0% accuracy on an exemplary MNIST inference task. Furthermore, a wavelength-multiplexed pJTC architecture shows potential for high throughput and energy efficiency, reaching 305 TOPS/W and 40.2 TOPS/mm2, based on currently available foundry processes. An efficient, compact, and low-latency convolution accelerator promises to advance next-generation AI capabilities across edge demands, high-performance computing, and cloud services. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_01117 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Near-energy-free Photonic Fourier Transformation for Convolution Operation Acceler Yang, Hangbo Peserico, Nicola Li, Shurui Ma, Xiaoxuan Schwartz, Russell L. T. Hosseini, Mostafa Babakhani, Aydin Wong, Chee Wei Gupta, Puneet Sorger, Volker J. Optics Computational Physics Convolutional operations are computationally intensive in artificial intelligence services, and their overhead in electronic hardware limits machine learning scaling. Here, we introduce a photonic joint transform correlator (pJTC) using a near-energy-free on-chip Fourier transformation to accelerate convolution operations. The pJTC reduces computational complexity for both convolution and cross-correlation from O(N4) to O(N2), where N2 is the input data size. Demonstrating functional Fourier transforms and convolution, this pJTC achieves 98.0% accuracy on an exemplary MNIST inference task. Furthermore, a wavelength-multiplexed pJTC architecture shows potential for high throughput and energy efficiency, reaching 305 TOPS/W and 40.2 TOPS/mm2, based on currently available foundry processes. An efficient, compact, and low-latency convolution accelerator promises to advance next-generation AI capabilities across edge demands, high-performance computing, and cloud services. |
| title | Near-energy-free Photonic Fourier Transformation for Convolution Operation Acceler |
| topic | Optics Computational Physics |
| url | https://arxiv.org/abs/2504.01117 |