Task splitting and mitigating learning prediction method based on multi-source domains

Abstract

The invention relates to a task splitting and mitigating learning prediction method based on multi-source domains. The goal of this method is to divide the domains involving the characteristic items of a target domain into corresponding sub-domains under the condition that a complex target task is difficult to train and learn with an extremely limited amount of target domain data so as to divide out the sub-tasks corresponding to the sub-domains. Through the domain characteristic amount corresponding to the sub-tasks, the sub-task model weights are calculated; through the use of the characteristic items required in the sub-domains corresponding to the sub-tasks, the characteristic mapping manner is utilized to establish the sharing characteristic space so that in ample multi-source domain sample, the sub-task models could be trained out; the associated characteristic item sharing parameter models are extracted to integrate initial target task models to be rapidly fit in combination with a gradient raising method to realize the task splitting and mitigating prediction. The invention serves as a cross-discipline model mitigating prediction method using the task splitting manner.

Description

technical field [0001] The present invention relates to the direction of transfer learning in the field of machine learning, in particular to a task splitting transfer learning prediction method based on multi-source domains. Background technique [0002] Traditional machine learning prediction methods do not get rid of the constraints of independent and identical distribution of training data and test data, so that machine learning methods must be executed under the assumption that sample data is independent and identically distributed. However, with the popularity of big data in real applications , this premise is difficult to adapt to the development of the era of data intelligence. Aiming at the problems that machine learning methods are not suitable for cross-domain learning due to the scarcity of training samples, and the collection and labeling of training samples are expensive, transfer learning, as a technical method of human-like learning, has received extensive at...

AI Technical Summary

Problems solved by technology

[0004] In view of the defects existing in the prior art, the purpose of the present invention is to propose a multi-source domain-based task splitting transfer learning prediction method to solve the problem of predicting complex target tasks associated with multiple fields, and to prevent and avoid cross-domain transfer learning. Occurrence of under-transfer and negative transfer, reduce the complexity of the model by task splitting method, solve the problem of over-fitting, and improve the generalization ability of the model

Images