Deep reinforcement learning-based incomplete information game method, device, system and storage medium

A reinforcement learning and incomplete technology, applied in the field of artificial intelligence, can solve problems such as complex tasks, lack of solutions, untimely acquisition of reward signals, etc., to achieve effective exploration and improve solution efficiency

Active Publication Date: 2019-11-01
HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
View PDF6 Cites 31 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for 3D incomplete information video games with huge state dimensions, complex tasks, a...

Method used

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
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep reinforcement learning-based incomplete information game method, device, system and storage medium
  • Deep reinforcement learning-based incomplete information game method, device, system and storage medium
  • Deep reinforcement learning-based incomplete information game method, device, system and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] The invention discloses an incomplete information game method based on deep reinforcement learning. The invention formulates a new reward signal supplementary mechanism through experiments and improvements on related deep reinforcement learning algorithms, and combines it with target detection technology to apply it to 3D incomplete information video game agent game algorithm, so that the machine can realize the decision-making process from perception to action and the ability of self-learning and exploration like human beings.

[0030] The present invention takes machine game and deep reinforcement learning algorithm as the main research contents, adopts the incomplete information 3D video game of Doom as the test platform of intelligent body game level, and studies the value model, strategy gradient, scope of application, efficiency problem and memory mechanism in reinforcement learning And the problem of sparse reward value is analyzed, and the limitations of related de...

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

PUM

No PUM Login to view more

Abstract

The invention provides a deep reinforcement learning-based incomplete information game method, a device, a system and a storage medium. The method comprises the steps of exploring and utilizing a mechanism to improve a strategy gradient algorithm, adding a memory unit into a deep reinforcement learning network, and optimizing a reward value by a self-driven mechanism. The beneficial effects of theinvention are that: the method is suitable for large-scale production. The problem of high variance frequently occurring in a strategy gradient algorithm is solved through a baseline function. For the problem of high time complexity in the reinforcement learning sampling and optimization process, a parallel mechanism is adopted to improve the model solving efficiency. Through a self-driven mechanism, an intelligent agent is helped to explore the environment more effectively while making up for the sparse environment reward value.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an incomplete information game method, device, system and storage medium based on deep reinforcement learning. Background technique [0002] In recent years, with the improvement of computer computing power and the development of big data technology, artificial intelligence has once again entered the golden age of development, and deep learning and reinforcement learning are the most eye-catching technologies in this peak of artificial intelligence development. Many scientific researchers and enterprises have increased their research on artificial intelligence. Countries around the world, including my country, have listed artificial intelligence research as an important development strategy at present, and artificial intelligence has even become the most important component of the comprehensive strength of countries in the world in the future. part. [000...

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

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06F18/214
Inventor 王轩漆舒汉蒋琳曹睿李明豪廖清李化乐张加佳刘洋夏文
Owner HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap