Adversarial-learning-based multi-source-domain adaptive migration method and system

Abstract

The invention discloses an adversarial-learning-based multi-source-domain adaptive migration method and system. The method comprises: step one, pre training is carried out by using all-source-domain data and a representation network and a classifier of a target model are initialized; step two, multi-path adversarial adversarial processing is carried out on multi-source-domain data and target-domain data and a representation network and a multi-path discriminator of the target model are updated; step three, adversarial scores between the source domains and the target domain are calculated; stepfour, target domain classification is carried out based on the classifiers and the adversarial scores of all source domains; step five, a target domain pseudo sample with a high confidence coefficient is selected for fine tuning of the representation network and the classifier of the target model; and step six, the steps from the step two to the step five are carried out again until model convergence is realized or a maximum iteration number of times is reached, and then training is stopped. According to the invention, reliance on the hypothesis of consistency of the single-source-domain tagset and the target domain is eliminated; and a negative migration phenomenon existing in the multi-source domain adaptation process is avoided effectively.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular to a multi-source domain adaptive migration method and system based on adversarial learning. Background technique [0002] With the continuous generation of large-scale data and the difficulty of relying on manual information labeling, domain adaptation transfer methods have gradually become a very important research topic in the field of machine learning. Domain adaptation learning aims to adapt the feature distribution between data in different domains, improve the performance of classifiers after migration between different domains, and solve the problem of lack of labeling information in target domain data. The domain adaptation transfer method is also a key technical means in the industry, and has important applications in many fields such as face recognition, automatic driving, and medical imaging. [0003] At present, most domain adaptation learning methods ...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the deficiencies in the above-mentioned prior art, the purpose of the present invention is to provide a multi-source domain adaptation migration method and system based on adversarial learning, so as to extend the existing single-source domain adaptation process based on adversarial learning to multi-source domain adaptation. Source domain adaptation no longer relies on the assumption that a single source domain label set is consistent with the target domain, and can effectively avoid the negative transfer phenomenon that exists in the process of multi-source domain adaptation

Images