Fictitious self-play-based multi-person incomplete information game policy resolving method, device and system as well as storage medium

Abstract

The invention provides a fictitious self-play-based multi-person incomplete information game policy resolving method, device and system as well as a storage medium. The method comprises the followingsteps: specific to a two-person gaming condition, implementing the generation of an average policy by using multi-type logistic regression and reservoir sampling, and implementing the generation of anoptimal response policy by using a DQN (Deep Q-Network) and annular buffering memory; and specific to a multi-person gaming condition, implementing the optimal response policy by using a multi-agentproximal policy optimization (MAPPO) algorithm, and meanwhile adjusting agent training by using multi-agent NFSP (Neural Fictitious Self-Play). The method has the beneficial effects that a fictitiousself-play algorithm framework is introduced; the Texas Poker policy optimizing process is partitioned into optimal response policy learning and average policy learning which are implemented by simulation learning and deep enhancement learning respectively; and a more universal multi-agent optimal policy learning method is designed.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, and in particular to a method, device, system and storage medium for solving a multiplayer incomplete information game strategy based on virtual self-play. Background technique [0002] Machine games and artificial intelligence are inextricably linked, which is an important aspect that reflects the development of artificial intelligence. Many well-known scholars in the computer field have conducted related research on machine games: the father of computer von Neumann and the mathematician Aumann proposed the minimax method in the game. Alan Turing, the father of artificial intelligence, provided the theoretical basis for developing computer chess programs. This theory was later used to design the world's first computer-programmed chess on ENIAC. For more than half a century, many major research achievements in the field of machine games are considered to be important mil...

AI Technical Summary

Problems solved by technology

[0005] Many decision-making problems in reality can be abstracted as strategy optimization problems of games with incomplete information, but the current strategy optimization algorithms for incomplete information, such as Master Lengpu, can only solve game problems with two people, discrete actions, and simple states , cannot be well applied in solving real-world decision-making problems

Images