Decision-making method based on deep reinforcement learning

A technology of reinforcement learning and decision-making methods, applied in the field of decision-making based on deep reinforcement learning, can solve problems such as slow training process, poor performance, and difficult to explain results, so as to avoid catastrophic decision-making and improve training effect
CN112295237APending Publication Date: 2021-02-02SHENZHEN UNIV
5 Cites 7 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENZHEN UNIV
Publication Date
2021-02-02

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a decision-making method based on deep reinforcement learning. The method comprises the steps that an intelligent agent makes a decision according to environment information and selects a decided action; the intelligent agent compares the decided action with a knowledge base, and decides to execute the decided action or replace the decided action based on a preset rule setin the knowledge base; the intelligent agent executes the decided action or the replaced action in the environment, obtains rewards and new environment information from the environment, combines the old environment information, the action, the rewards and the new environment information into experience information, and stores the experience information into an experience playback pool; and a preset amount of experience information is randomly selected from the experience playback pool to update the deep reinforcement learning model so as to guide the next iteration. By utilizing the method, the training time can be shortened, disastrous decisions can be avoided, and the method can be widely applied to the field of dynamic decisions.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present invention relates to the field of artificial intelligence, and more specifically, to a decision-making method based on deep reinforcement learning. Background technique

[0002] Reinforcement learning is a field in machine learning that describes and solves problems in which agents learn strategies to maximize rewards or achieve specific goals during interactions with the environment.

[0003] Currently, deep reinforcement learning has been successfully applied to a variety of dynamic decision-making domains, especially those with large state spaces. However, deep reinforcement learning also faces some problems. First, its training process can be very slow and resource-intensive, and the final system is usually fragile, the results are difficult to interpret, and it performs poorly for a long time at the beginning of training. Moreover, for applications in robotics and critical decision support systems, it is even possible to make catastro...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More