A hierarchical network structure search method, device and readable storage medium

Abstract

The present invention provides a hierarchical network structure search method, device and readable storage medium, comprising: S1, constructing a supernetwork; S2, acquiring image data and using them as training data of network parameters and structural parameters of the supernetwork respectively; S3, generating Feature map, calculate the cross-entropy loss function of the network parameters, update the network parameters of the super network; S4, generate the feature map and KL divergence loss function, calculate the cross-entropy loss function of the structural parameters, and obtain the semi-implicit variational discarding loss function, Training updates the structural parameters of the super network and obtains the discarding probability; S5, uses the discarding probability to update the basic unit, and updates the annealing parameters; repeats S3-S5, updates the network parameters and structural parameters; S6, obtains the final network. The invention greatly reduces search time and calculation complexity while ensuring high performance, ensures search stability and practicability, can be used in the fields of image target detection and classification, and improves image processing speed.

Description

technical field [0001] The invention relates to the technical fields of artificial intelligence and image processing, in particular to a hierarchical network structure search method, computer equipment and a readable storage medium thereof, and the application of the method in image target detection, semantic segmentation and classification. Background technique [0002] With the development of computing power and deep learning, the development of artificial intelligence is getting faster and faster. In the initial image processing, due to the low quality of the collected images, the requirements for feature extraction are relatively low, so the main technology is manual feature extraction. Later, the image quality continued to improve, and the accuracy requirements for image processing continued to increase. Some statistical pattern recognition methods such as SVM and signal processing methods such as wavelet transform have made some progress in the development of image pro...

AI Technical Summary

Problems solved by technology

Secondly, the existing search method based on the DARTS search space will collapse due to the excessive number of search epochs when searching the network. This collapse phenomenon is reflected in the fact that the final structure becomes a fully connected neural network without parameters. Completely incompetent on image classification tasks

And this collapse phenomenon also makes when the network structure search is directly used for image classification or other downstream tasks, because it is impossible to determine when the network structure converges, resulting in insufficient performance of downstream tasks

Images