Check patentability & draft patents in minutes with Patsnap Eureka AI!

A conversion method for model deduction to intelligent deduction

A model and intelligent technology, applied in inference methods, neural learning methods, biological neural network models, etc., can solve problems such as the use of deep reinforcement learning without a detailed introduction

Active Publication Date: 2022-05-27
NAT UNIV OF DEFENSE TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Patent No.: 201811313124.X proposes an unsupervised intelligent confrontation deduction system based on deep reinforcement learning. This method does not introduce in detail how deep reinforcement learning is used in the confrontation deduction system

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
  • A conversion method for model deduction to intelligent deduction
  • A conversion method for model deduction to intelligent deduction
  • A conversion method for model deduction to intelligent deduction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The present invention is further described below in conjunction with the accompanying drawings, but the present invention is not limited in any way, any transformation or substitution based on the teachings of the present invention, is within the scope of the present invention.

[0031] Object of the present invention is to convert the equipment unit model data of the adversarial two sides, the unit behavior model data input model into an image model that can be used for deep reinforcement learning, the unit behavior decision-making output into an output that can be used for deep reinforcement learning, the virtual environment deduction image model transformation, to adapt to the needs of deep reinforcement learning, to provide a basis for intelligent adversarial unit deduction. Adversarial units of the present invention include ground adversarial units, aerial adversarial units and surface adversarial units.

[0032] as Figure 1 As shown, the present invention discloses a ...

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 discloses a transformation method for model deduction to intelligent deduction, which converts the digital input of the model of the confrontation unit in the model deduction, the digital input of the rule behavior and the digital output of the behavior decision of the confrontation unit into the input and input of the image model in the intelligent deduction. Output: Carry out discrete sampling of the movement process and grouping process of the countermeasure unit in the virtual environment deduction in the intelligent deduction, and convert it into a sequence image of the graphical model; transform the damage assessment of the virtual environment deduction into each unit of the image model in the intelligent deduction The damage and overall damage assessment; the input of the image model in the intelligent deduction is used as the input of the training and testing model of deep reinforcement learning, and the training and testing model of deep reinforcement learning outputs action and strategy value vectors. The invention converts the unit behavior decision output into an output that can be used for deep reinforcement learning, so as to meet the requirements of deep reinforcement learning and provide a basis for intelligent deduction.

Description

Technical field [0001] The present invention belongs to the field of adversarial deduction techniques, in particular to a method for converting model deduction to intelligent deduction. Background [0002] Intelligent confrontation refers to the use of intelligent weapons and means on the basis of information confrontation to achieve a high-tech confrontation form with efficient command, precision strikes, automated operations, and intelligent behavior as the main purpose. In essence, intelligent confrontation is the radiation and extension of human intelligence to the information confrontation site and confrontation equipment. From the perspective of confrontation procedures and means, intelligent confrontation includes intelligent command and control and offensive and defensive confrontation of intelligent equipment. The main symbol of intelligent confrontation is the emergence of intelligent confrontation equipment groups and intelligent countermeasures. [0003] Confrontation...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06F30/20G06N5/04G06N3/08G06N3/04
CPCG06F30/20G06N5/04G06N3/08G06N3/045
Inventor 曾向荣钟志伟刘衍张政
Owner NAT UNIV OF DEFENSE TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More