Meta-invariant feature space learning method for cross-domain prediction

Abstract

The invention relates to a meta-invariant feature space learning method for cross-domain prediction, which is characterized by comprising the following steps of: taking existing data as source domain data, and grouping and pairing the source domain data; establishing a prediction model for the data in each pair, then constructing an invariant feature space learning model of the paired data, and learning the invariant feature space of each pair of data through cooperative training; and taking the invariant feature space learning model as a base model, learning meta-invariant feature spaces between different pairs through a meta learning method to obtain a meta-invariant feature space learning model, and then predicting a target domain based on the meta-invariant feature space learning model. According to the method, invariant feature spaces under two working conditions are obtained through a collaborative learning mode, and the problem of edge distribution adaptation is solved. The cross-domain prediction precision is improved.

Description

technical field [0001] The present invention relates to the field of artificial intelligence, in particular to a cross-domain prediction method oriented to intelligent manufacturing, in particular to a meta-invariant feature space learning method for cross-domain prediction. Background technique [0002] Cross-domain prediction is an important research issue in the field of machine learning. In the field of manufacturing, the main reason is that due to the large changes in working conditions, the marginal distribution and conditional distribution of data are quite different, and it is difficult to adapt the model trained on the source domain to new conditions. Working conditions, a typical problem in the manufacturing field is tool wear prediction. Real-time monitoring of tool wear is of great significance to the dynamic control of the machining process, especially for complex aircraft parts that use a large number of difficult-to-machine materials. The tool wear during the ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to address the problem of cross-domain prediction, and propose a meta-invariant feature space learning method, which takes the invariant feature space learning model as the base model, learns the meta-invariant feature space through the meta-learning method, and then based on the meta-invariant Variational Feature Space Learning Models for Prediction of Target Domains

Images