.png){kind=icon}
.png){kind=icon}
Return
Coupling Enhanced Diffractive Deep Neural Network with Structural Nonlinearity
.png){kind=icon}
OA
AI
DOI：10.1002/adpr.202500038.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing complexity of deep learning models poses stringent requirements on electronic computers. Diffractive deep neural networks (D2NNs), as one of the most representative optical computing architectures, have emerged as a significant substitute for electronic-based devices due to the advantages of high speed, low power consumption, and high parallelism. However, the absence of optical nonlinearity constrains the potential advancement of D2NNs. Recent progress in structural nonlinearity has offered a promising avenue for addressing this issue, but it necessitates complex digital data pre-encoding. Herein, structural nonlinearity is introduced into D2NNs by incorporating encoding-free data repetition layers, enabling high-order optical nonlinearity while reducing the system complexity. The effectiveness of different data repetition manners demonstrates the robustness of this approach. Additionally, to enhance the design accuracy of D2NNs, a graph neural network framework is developed to characterize the coupling effects in metasurface layers and integrate it into D2NNs. This work provides a novel approach for the design of optical computing devices and holds significant importance for the development of high-performance and highly integrated all-optical devices.
Keywords:
coupling
diffractive deep neural network
nonlinearity
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
18.8
Papers:
981
Citations:
3.6K