A man-machine conversation method based on actor-critic reinforcement learning algorithm in cyclic network

Abstract

The invention relates to a circulating network man-machine conversation method based on an actor critic reinforcement learning algorithm. The system consists of two subsystems: dialogue generation system and emotion analysis system. The session generation system is based on the gate loop unit network model and uses the tagged session data set for training. Furthermore, we optimize the parameters of the trained model using the actor-critic algorithm in reinforcement learning. That is, we use the trained model to build two networks, called the 'actor' network and the 'critic' network; Further, in order to reduce training time and improve resource utilization, we have created multiple processes that assign a pair of 'actors' and 'critics' to each process.

Description

technical field

[0001] The invention relates to a man-machine dialogue method, in particular to a recurrent network man-machine dialogue method based on actor-critic reinforcement learning algorithm. Background technique

[0002] With the development of science and technology, new achievements are constantly emerging in the field of natural language processing, among which the development of human-computer dialogue technology is particularly eye-catching. At present, such as in the field of e-commerce, people communicate with each other mainly through online chatting. For store owners, most of the inquiries and after-sales information from customers are repetitive. In addition, for some larger shops, shop owners often need to hire customer service, which will undoubtedly increase operating costs. For customers, they hope that the shopkeeper can reply as soon as possible and decently. The human-computer dialogue system can complete the work of customer service, reduce the ...

Images