Neural network structure automatic search method based on automatic machine learning
A technology of machine learning and network structure, which is applied in the field of neural network, can solve problems such as the depth interval of the search process, performance loss, and too long search time of the neural network structure, and achieve performance loss control, search time reduction, and strong generalization ability Effect
Pending Publication Date: 2021-04-09
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF0 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The purpose of the present invention is to solve the problems of too long search time of neural network structure, depth interval of search process and performance loss, and proposes an automatic search method of neural network structure based on automatic machine learning
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
[0045] Embodiments of the present invention will be further described below in conjunction with the accompanying drawings.
[0046] Before describing the specific embodiments of the present invention, in order to make the scheme of the present invention more clear and complete, at first the abbreviations and key term definitions that appear in the present invention are explained:
[0047] dil_conv_3x3 and dil_conv_5x5: Both are candidate operations in depthwise separable convolution.
[0048] Such as figure 1 As shown, the present invention provides a kind of neural network structure automatic search method based on automatic machine learning, comprises the following steps:
[0049] S1: In the stage of model generation based on automatic machine learning, the search space S is determined by using the unit cell of the neural network;
[0050] S2: Use the search strategy to select a specific structure s from the search space S;
[0051] S3: Based on the search space S, use th...
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 neural network structure automatic search method based on automatic machine learning, and the method comprises the following steps: S1, determining a search space S based on a unit cell of a neural network; S2, selecting a specific structure s from the search space S by utilizing a search strategy; and S3, based on the search space S, searching the specific structure s by using the search strategy, and returning a search result to the search strategy to complete automatic search of the neural network structure. The search method is high in generalization ability, can be applied to various scenes such as computer vision and natural language processing, particularly has an obvious application effect for the intelligent customer service industry in the financial field, can obviously reduce the time cost in the modeling process, and improves the modeling efficiency.
Description
technical field [0001] The invention belongs to the technical field of neural networks, and in particular relates to an automatic search method for neural network structures based on automatic machine learning. Background technique [0002] With the rapid development of artificial intelligence technology, machine learning is widely used in the fields of natural language processing and computer vision, and the structure of the corresponding neural network is also becoming more complicated. In the field of deep learning, the neural network structure usually needs to be manually designed. In order to obtain an ideal neural network structure, it will consume a lot of manpower and material resources. Therefore, in order to liberate manpower, the research on automatic search of neural network structure has gradually been paid more attention by researchers. Based on the above situation, the present invention proposes an automatic search method for neural network structure based on...
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): G06N3/08G06N3/04G06N20/00
CPCG06N3/08G06N3/084G06N20/00G06N3/047
Inventor 陈波史特左御丁王庆先
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA