Method for quickly mixing high-order attention domain adversarial network based on transfer learning

Abstract

The invention relates to a method for quickly mixing a high-order attention domain adversarial network based on transfer learning. The method comprises the steps of designing a quick mixing high-orderattention and domain adversarial adaptive network for a to-be-processed image data set; preprocessing the source domain and the target domain; importing the preprocessed source domain and the preprocessed target domain into the designed network in batches in sequence, obtaining weighted feature maps through a fast mixing high-order attention network, then inputting the weighted fine feature mapsinto a domain adversarial adaptive network for training, and finally performing probability operation through a full connection layer; respectively calculating average image classification accuracy ofthe source domain and the target domain; enabling a gradient inversion layer in back propagation to take a reverse gradient direction to form adversarial training, then performing iterative training,and applying a fast mixed high-order attention and domain adversarial adaptive network trained on a source domain directly to a target domain to carry out image classification. According to the invention, the recognition rate and migration capability of the unsupervised domain adaptive network in migration learning are improved.

Description

technical field [0001] The invention relates to a method for a fast mixed high-order attention domain confrontation network based on migration learning, and belongs to the technical field of neural network deep learning. Background technique [0002] So far, supervised learning methods employing fully annotated data for model training have achieved great success and have been successfully applied in many practical applications, such as image recognition, speech recognition, etc. But collecting enough training data in real-world scenarios is often expensive, time-consuming, and involves a lot of human resources or even impractical. A common solution to this problem is to leverage the rich knowledge in easily labeled source domains to facilitate efficient model learning in multiple label-scarce target domains, known as domain adaptation. Typically, domain adaptation includes supervised adaptation (in which a small amount of labeled target data is available for training) and u...

AI Technical Summary

Problems solved by technology

Although domain-adaptive transfer attention takes into account the difference in transferability of different images and explores images that are more similar across domains, these commonly used attention methods (i.e., spatial and channel attention) are based on first-order spatial distribution discriminative masks, Limited to mining simple and rough information, they are insufficient to capture complex high-order saliency information

Images