倾斜模式全光学衍射深度神经网络
Mingzhu Song1,2, Xuhui Zhuang1,2, Lu Rong3
1Liaoning Key Laboratory of Marine Sensing and Intelligent Detection, Dalian Maritime University, Dalian 116026, China.
Micromachines
|January 25, 2025
概括
倾斜模式衍射深度神经网络 (T-D2NN) 消除了传统D2NN中光反射引起的性能问题. 与传统设计相比,这些T-D2NN显示出更高的分类结果,即使是那些考虑到反射的设计.
科学领域:
- 光学是什么?光学是什么?光学是什么?
- 人工智能的人工智能
- 光子学 是一个光子学.
背景情况:
- 衍射深度神经网络 (D2NN) 使用级联衍射面具进行计算.
- 在D2NN中的面具之间的层间反射降低了网络性能,并且很难模拟.
- 高指数衍射面罩加剧了由于反射而导致的性能退化.
研究的目的:
- 为了消除D2NN中层间反射引起的性能降低.
- 引入一种新的架构:倾斜模式全光学衍射深度神经网络 (T-D2NNs).
- 开发一个在倾斜模式中衍射传播的理论模型.
主要方法:
- 设计和模拟倾斜模式全光学衍射深度神经网络 (T-D2NNs).
- 在倾斜模式下开发了衍射传播的理论模型.
- 将T-D2NN与传统D2NN的性能进行比较,包括那些考虑层间反射的D2NN.
主要成果:
- T-D2NNs有效地消除了由层间反射引起的性能降低.
- 提出的理论模型准确地描述了倾斜模式中的衍射.
- 与传统的D2NN相比,T-D2NN在分类任务中取得了更好的分类结果.
结论:
- T-D2NNs为衍射神经网络中的层间反射问题提供了解决方案.
- 倾斜模式架构增强了全光衍射网络的推断能力.
- 这项工作推动了衍射深度神经网络的设计和应用.
相关概念视频
01:25Neural Circuits
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
01:25Maximizing the Directional Derivative
The directional derivative is a central concept in multivariable calculus that describes how a function changes at a given point when moving in a specified direction. This direction is represented by a unit vector, ensuring that only the orientation influences the rate of change. By varying the direction, different rates of change can be observed, demonstrating that the directional derivative depends strongly on the chosen direction.The directional derivative is computed using the gradient...