Model training method, object control method and device based on interactive application

Abstract

The present application discloses a model training method applied in the field of artificial intelligence, including: obtaining the first state information corresponding to the first object and the second state information corresponding to the second object; based on the first state information, through the first The engagement network model to be trained acquires the first action information; based on the second state information, the second action information is acquired through the second engagement network model to be trained; based on the first state information, the second state information, the first action information and the second action The value expectation is obtained by discriminating the network model; the model parameters of the first engagement network model to be trained are updated according to the value expectation to obtain the first engagement network model. The application also discloses an interactive application-based object control method and device. The application can realize the cooperation and cooperation between multiple non-player characters, reduce the operation difference between non-player characters and real players, and improve the level of fighting between non-player characters and real players.

Description

technical field [0001] The present application relates to the field of artificial intelligence, in particular to a model training method, an object control method and device based on an interactive application. Background technique [0002] Turn-based combat is a combat mode in the game, that is, multiple combat characters belonging to the same party activate their skills in turn, and the two combatants take turns to attack. If the current combat character of one of the parties dies, it will be the turn of another combat character of the defeated party to fight in the next round, and so on, until all the combat characters of one party are killed, the turn-based battle will not end. [0003] Currently, one party is a real player in a turn-based battle, while the other party can be a non-player character (Non-PlayerCharacter, NPC). NPC mainly launches an attack by establishing a decision tree. The decision tree can establish a tree structure by selecting relevant features, an...

AI Technical Summary

Problems solved by technology

[0004] However, for the method of building a decision tree, due to the small amount of sample data, the action strategy that NPC can produce is relatively simple, and often lacks cooperation and cooperation. Improve the level of NPC and real players in the process of fighting

Images