GRU network model based on ordinary differential equation, and feature extraction method and device
A technology of ordinary differential equations and network models, applied in the field of GRU network model and feature extraction based on ordinary differential equations, can solve problems such as large memory usage, and achieve the effects of reducing memory usage, reducing storage space, and improving memory efficiency
Pending Publication Date: 2021-12-14
THE SECOND AFFILIATED HOSPITAL ARMY MEDICAL UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The essence of the neural network is mainly to fit a complex composite function. The number of composites is the number of layers of the neural network. To find the gradient of the parameters, it is easy to think of the chain rule. However, it is necessary to retain all layers during forward propagation. Activation values, and use these activation values in backpropagation, which is very memory intensive and is a big bottleneck for the training process of deep models
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0029] The present invention will be further elaborated below in conjunction with the accompanying drawings and specific embodiments. It should be noted here that although the descriptions of these embodiments are used to help the understanding of the present invention, they are not intended to limit the present invention. Specific structural and functional details disclosed herein are for purposes of describing example embodiments of the invention only. However, the invention may be embodied in many alternative forms and should not be construed as limited to the embodiments set forth herein.
[0030] It will be understood that, although the terms first, second etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing f...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a GRU network model based on an ordinary differential equation, and a feature extraction method and device, the GRU network model comprises at least one gating loop unit, and each gating loop unit in the at least one gating loop unit comprises a candidate hidden layer. And the candidate hidden layer calculates input information of the candidate hidden layer based on an ordinary differential equation and converts the input information into a hidden state sequence for output. The GRU network model does not need to consume a large amount of space to store intermediate results in the training process, the storage space is reduced, and the memory efficiency is improved.
Description
technical field [0001] The invention relates to the technical field of deep learning, in particular to a GRU network model based on ordinary differential equations and a feature extraction method and device. Background technique [0002] In recent years, deep learning has developed rapidly, especially because of the sensation of Google's AlphaGo some time ago, and the domestic research boom of this technology has also begun. Deep learning is still in the development stage. Researchers have applied deep learning to related image processing and computer vision fields. Deep learning algorithms have achieved outstanding results in many supervised learning problems. The performance of such aspects is far superior to traditional machine learning algorithms, and some even surpass the human level. At present, the research interests of deep learning researchers have gradually shifted from supervised learning to reinforcement learning, semi-supervised learning, and unsupervised learn...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06N3/04G06N3/08G06F17/13G06K9/46G06K9/62
CPCG06N3/08G06F17/13G06N3/044G06F18/24
Inventor 宗兆文黄军建蒋仁庆钟鑫杜文琼贾益君周小林杨昊洋
Owner THE SECOND AFFILIATED HOSPITAL ARMY MEDICAL UNIV