Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2308.14015 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917602103656448 |
|---|---|
| author | Liu, Yi-Feng Ren, Rui-Yao Hou, Dai-Bao Weng, Hai-Zhong Wang, Bo-Wen Huang, Ke-Jie Lin, Xing Liu, Feng Li, Chen-Hui Jin, Chao-Yuan |
| author_facet | Liu, Yi-Feng Ren, Rui-Yao Hou, Dai-Bao Weng, Hai-Zhong Wang, Bo-Wen Huang, Ke-Jie Lin, Xing Liu, Feng Li, Chen-Hui Jin, Chao-Yuan |
| contents | Due to their intrinsic capabilities on parallel signal processing, optical neural networks (ONNs) have attracted extensive interests recently as a potential alternative to electronic artificial neural networks (ANNs) with reduced power consumption and low latency. Preliminary confirmation of the parallelism in optical computing has been widely done by applying the technology of wavelength division multiplexing (WDM) in the linear transformation part of neural networks. However, inter-channel crosstalk has obstructed WDM technologies to be deployed in nonlinear activation in ONNs. Here, we propose a universal WDM structure called multiplexed neuron sets (MNS) which apply WDM technologies to optical neurons and enable ONNs to be further compressed. A corresponding back-propagation (BP) training algorithm is proposed to alleviate or even cancel the influence of inter-channel crosstalk on MNS-based WDM-ONNs. For simplicity, semiconductor optical amplifiers (SOAs) are employed as an example of MNS to construct a WDM-ONN trained with the new algorithm. The result shows that the combination of MNS and the corresponding BP training algorithm significantly downsize the system and improve the energy efficiency to tens of times while giving similar performance to traditional ONNs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2308_14015 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Slimmed optical neural networks with multiplexed neuron sets and a corresponding backpropagation training algorithm Liu, Yi-Feng Ren, Rui-Yao Hou, Dai-Bao Weng, Hai-Zhong Wang, Bo-Wen Huang, Ke-Jie Lin, Xing Liu, Feng Li, Chen-Hui Jin, Chao-Yuan Signal Processing Hardware Architecture Due to their intrinsic capabilities on parallel signal processing, optical neural networks (ONNs) have attracted extensive interests recently as a potential alternative to electronic artificial neural networks (ANNs) with reduced power consumption and low latency. Preliminary confirmation of the parallelism in optical computing has been widely done by applying the technology of wavelength division multiplexing (WDM) in the linear transformation part of neural networks. However, inter-channel crosstalk has obstructed WDM technologies to be deployed in nonlinear activation in ONNs. Here, we propose a universal WDM structure called multiplexed neuron sets (MNS) which apply WDM technologies to optical neurons and enable ONNs to be further compressed. A corresponding back-propagation (BP) training algorithm is proposed to alleviate or even cancel the influence of inter-channel crosstalk on MNS-based WDM-ONNs. For simplicity, semiconductor optical amplifiers (SOAs) are employed as an example of MNS to construct a WDM-ONN trained with the new algorithm. The result shows that the combination of MNS and the corresponding BP training algorithm significantly downsize the system and improve the energy efficiency to tens of times while giving similar performance to traditional ONNs. |
| title | Slimmed optical neural networks with multiplexed neuron sets and a corresponding backpropagation training algorithm |
| topic | Signal Processing Hardware Architecture |
| url | https://arxiv.org/abs/2308.14015 |