Neural network search method, image processing method and device, equipment and medium

A technology of neural network and search method, applied in the direction of biological neural network model, neural architecture, still image data index, etc., can solve the problems that the performance of neural network needs to be further improved, and achieve the effect of improving the degree of freedom of network structure and improving performance

Active Publication Date: 2021-10-29
BEIJING SENSETIME TECH DEV CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In some network structure search methods, several candidate block structures are pre-defined, and each layer of the network can only choose among these predefined block structures, and the performance of the obtained neural network needs to be further improved.

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Neural network search method, image processing method and device, equipment and medium
  • Neural network search method, image processing method and device, equipment and medium
  • Neural network search method, image processing method and device, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the specific technical solutions of the invention will be further described in detail below in conjunction with the drawings in the embodiments of the present application. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0041] The embodiment of the present application proposes a gradient-based neural network search scheme, which simultaneously optimizes the overall structure of the network and the internal implementation of the unit, and uses the multi-branch search space to search the network layer by layer.

[0042] The technical solutions provided by the embodiments of the present disclosure are mainly applied to the search of convolutional neural networks, but may also be applied to the searches of other types of neural networks, which is not limited in the embo...

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

No PUM Login to view more

Abstract

Embodiments of the present application provide a neural network search method, image processing method, device, device, and medium, in which multiple optional channel numbers corresponding to multiple optional convolution operations are sampled to obtain the first A set of candidate channel numbers of the network unit; based on at least two candidate channel numbers included in the candidate channel number set of the first network unit, the plurality of optional convolution operations are sampled to obtain the first network unit At least two sets of candidate convolution operations; adjusting network parameters of the neural network based on an image processing result obtained by processing an input image through a plurality of network units of the neural network.

Description

technical field [0001] The embodiments of the present application relate to the field of machine learning, involving but not limited to neural network search methods, image processing methods and devices, devices and media. Background technique [0002] The design of convolutional neural networks is the fundamental work of all deep learning related methods. A good neural network structure can improve the performance of deep learning tasks. In some network structure search methods, several candidate block structures are pre-defined, and each layer of the network can only choose from these predefined block structures, and the performance of the obtained neural network needs to be further improved. Contents of the invention [0003] Embodiments of the present application provide a neural network search solution and an image processing solution. [0004] An embodiment of the present application provides a neural network search method, the method comprising: [0005] Samplin...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/04G06F16/51G06F16/58
CPCG06F16/51G06F16/58G06N3/045
Inventor 次元政林宸武伟
Owner BEIJING SENSETIME TECH DEV CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap