Question and answer model training method, computer equipment and readable storage medium

Abstract

The invention discloses a question and answer model training method, computer equipment and a computer readable storage medium. The question and answer model training method comprises the steps of establishing a text matching model and a meta-learning model, wherein the text matching model and the meta-learning model have the same deep neural network structure; obtaining a test task and a plurality of different training tasks; updating meta-network parameters of the meta-learning model based on respective first matching losses of the plurality of different training tasks, wherein the first matching losses are used for representing matching errors of respective text matching models of the plurality of different training tasks; and generating initial parameters of a text matching model by utilizing the updated meta-network parameters, training the text matching model by using the test task to update the initial parameters of the text matching model, and taking the text matching model obtained after training as a question and answer model. By means of the mode, the question and answer model can be applied to the question and answer field with few samples, and the question and answer pair matching accuracy is improved.

Description

technical field [0001] The present application relates to the technical field of machine learning, and in particular, to a training method of a question answering model, a computer device, and a computer-readable storage medium. Background technique [0002] Question answering system is an important branch of information retrieval and natural language processing. In recent years, with the rise of the Internet, user-generated content has accumulated a lot on the network, resulting in rich open-domain question-answer and vertical-domain question-and-answer data. [0003] In the community question answering scenario, a common application is to select the answer that best matches the current user question among the candidate answers, that is, answer sorting and selection. A question answering model with high accuracy can quickly locate the answer that best matches the new question among the candidate answers, improve the question answer rate, and improve the user experience, but...

AI Technical Summary

Problems solved by technology

[0005] The technical problem mainly solved by this application is to provide a question answering model training method, computer equipment and computer-readable storage medium, which can solve the problem of difficult model training due to small sample size

Images