Graph pooling method and apparatus, equipment, and storage medium
A pooling and subgraph technology, applied in the field of image processing, can solve the problems of large scale of neural network model, large amount of computational data, low efficiency of graph pooling, etc.
Pending Publication Date: 2020-12-25
TENCENT TECH (SHENZHEN) CO LTD
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
In the existing technology, graph pooling is to reduce the size of the entire graph from the topological level to obtain a more compact representation. Using the graph neural network model to obtain the weight of node filtering is compared to only using node features to map to obtain weights. Considering The interaction between nodes requires a large-scale neural network model, a large amount of calculation data, and low efficiency of graph pooling
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
[0052] The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.
[0053] See figure 1 , is a schematic structural diagram of the network architecture provided by the embodiment of the present invention. Such as figure 1 As shown, the network architecture may include a cloud server 2000 and a user terminal cluster; the user terminal cluster may include multiple user terminals, such as figure 1 As shown, it specifically includes user terminal 3000a, user terminal 3000b, ..., user terminal 3000n; as figure 1 As shown, the user terminal 3000a, the ...
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 embodiment of the invention discloses a graph pooling method and apparatus, equipment and a storage medium. The method comprises the steps that: a to-be-pooled graph feature representation is obtained, the to-be-pooled graph feature representation is segmented into a plurality of sub-graph feature representations, wherein one sub-graph feature representation comprises one or more dimensions ofgraph feature representations in the to-be-pooled graph feature representation; a graph attention output feature corresponding to each sub-graph feature representation is determined based on a graphneural network model, wherein one sub-graph feature representation corresponds to one graph attention output feature; graph attention output features corresponding to the sub-graph feature representations are aggregated, so that to-be-screened graph feature representations can be obtained; and target graph features are screened out according to the feature representations of a plurality of to-be-screened objects included in the to-be-screened graph feature representation and the weight of each to-be-screened object, and a pooled target graph feature representation is determined based on the target graph features. By adopting the embodiment of the invention, a picture can be pooled after the picture features are segmented according to the dimensions, and the pooling efficiency of the picture is improved.
Description
technical field [0001] The present application relates to the technical field of image processing, and in particular to a graph pooling method, device, equipment, and storage medium. Background technique [0002] Pooling is a method of reducing the feature representation size of the original input and representing the existing input with fewer feature values. The purpose of graph pooling is to reduce the representation size of the original input graph, usually by reducing the feature dimension or deleting some nodes and edges. In the existing technology, graph pooling is to reduce the size of the entire graph from the topological level to obtain a more compact representation. Using the graph neural network model to obtain the weight of node filtering is compared to only using node features to map to obtain weights. Considering The mutual influence between nodes requires a large-scale neural network model, a large amount of calculation data, and low efficiency of graph pooli...
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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/213
Inventor 曹雨方蒙
Owner TENCENT TECH (SHENZHEN) CO LTD