Clustering method and device based on variational auto-encoder
An autoencoder and clustering method technology, which is applied in the field of clustering methods and devices based on variational autoencoders, and can solve the problems of low clustering accuracy and poor clustering performance.
Pending Publication Date: 2021-03-12
TSINGHUA UNIV
View PDF0 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, LTVAE uses a complex tree model to iteratively optimize the structural relationship between its hidden variables, and its clustering performance is poor. GMVAE and VaDE use mean field approximation, but this approximation may not hold true in actual data, resulting in poor clustering accuracy. lower
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
[0015] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary, and are intended to explain the present application, and should not be construed as limiting the present application.
[0016] The variational autoencoder-based clustering method and device according to the embodiments of the present application will be described below with reference to the accompanying drawings. It should be noted that the implementation subject of the variational autoencoder-based clustering method in the embodiment of the present application is a variational autoencoder-based clustering device, which can be applied to any In a computer device, the computer device can perform a clustering function based on a v...
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 clustering method and device based on a variational auto-encoder, and the method comprises the steps of obtaining input data x; combining the input data x with a variationalauto-encoder to obtain a corresponding discrete category hidden variable y and a continuous Gaussian hidden variable z; obtaining corresponding joint probability distribution q (z, y|x) by combining the input data x, the discrete category implicit variable y and the continuous Gaussian implicit variable z with a variation auto-encoder; and directly decomposing the joint probability distribution q(z, y|x) by using a variational auto-encoder, and determining the category of the input data x according to the probability distribution of the discrete category implicit variable y. According to themethod, improvement is carried out on the basis of a variational auto-encoder, average field approximation is abandoned, probability distribution of discrete category implicit variables is obtained byadopting direct decomposition joint probability distribution, then the category of input data is determined, and the clustering effect and accuracy are improved.
Description
[0001] Cross References to Related Applications [0002] This application claims the priority of the Chinese patent application number "202011165489.X" submitted by Tsinghua University on October 27, 2020, with the application name "Clustering method and device based on variational autoencoder". technical field [0003] The present application relates to the technical field of deep learning, in particular to a clustering method and device based on a variational autoencoder. Background technique [0004] At present, unsupervised clustering is widely used in many practical fields, for example, the clustering of images or image features. In related technologies, data clustering is realized through the variational autoencoder model. For example, LTVAE greedily learns the tree structure between hidden variables by using the tree model on the hidden variable, and the complex prior constrained variational autoencoder constructed accordingly The encoder model is used for clustering...
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
CPCG06N3/08G06N3/045G06F18/2321
Inventor 裴丹李之涵
Owner TSINGHUA UNIV