Monte Carlo tree search and convolutional neural network-based Landlords strategy research method

A technology of convolutional neural network and Monte Carlo method, which is applied in the field of strategy research based on Monte Carlo tree search and convolutional neural network.

Pending Publication Date: 2020-10-02
GUIZHOU UNIV
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

On the contrary, there is less research on "Fighting the Landlord" now, mainly because it is difficult and not enough attention is paid to it

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
  • Monte Carlo tree search and convolutional neural network-based Landlords strategy research method
  • Monte Carlo tree search and convolutional neural network-based Landlords strategy research method
  • Monte Carlo tree search and convolutional neural network-based Landlords strategy research method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0041] see figure 1 , the present invention provides a technical solution: a research method based on Monte Carlo tree search and convolutional neural network fighting landlord strategy, comprising the following steps:

[0042] Randomly start the game and when each player plays a card (decision), take the player's current state (each player has played cards, the number of cards in the player's hand, the current player's hand, etc.) action as a direct child node of...

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 Monte Carlo tree search and convolutional neural network-based Landlords strategy research method in the technical field of machine learning. The method comprises the following steps that a game is started randomly, and when each player plays cards, the current state of the player serves as a root node, and actions possibly adopted by the player according to the Landlordsrule serve as direct child nodes of the root node; starting from a root node of the game tree, a Monte Carlo tree search algorithm is used to carry out continuous simulation sampling learning; when enough data is obtained by using the Monte Carlo tree search algorithm, a convolutional neural network CNN learning network is continuously trained by taking (states and possible playing cards and earnings corresponding to the possible playing cards in the current state) as data samples until the network is stable; for possible errors of the CNN network during learning, a strategy improvement algorithm is further used to correct and improve the learning result of the CNN network.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a research method for fighting landlord strategies based on Monte Carlo tree search and convolutional neural networks. Background technique [0002] In recent years, with the development of machine learning, this method has also achieved remarkable results in complete information games. The milestones are: On March 15, 2016, AlphaGo developed by Google, using methods such as deep learning and reinforcement learning, defeated the world Go champion Lee Sedol in the field of Go, and its iconic machine achieved superhuman performance in the field of Go. The subsequent training of AlphaGo Zero was carried out entirely through self-learning. AlphaGo Zero started from a random game without any supervision or human data, and only used black and white pieces on the board as the original input features, and finally won by a significant advantage through continuous learning. Alpha...

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 Applications(China)
IPC IPC(8): A63F13/46A63F13/822G06N3/04G06N3/08
CPCA63F13/46A63F13/822G06N3/08A63F2300/61A63F2300/807G06N3/045
Inventor 王以松彭啟文
Owner GUIZHOU UNIV
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