Decision-making method and system based on cloud platform

Abstract

The invention belongs to the technical field of artificial intelligence, and discloses a cloud platform-based decision-making method and system. The method comprises the steps of receiving a simulation training request; configuring hardware resources, and initializing a simulation engine, a decision engine and a learning engine; using the decision engine to obtain an instruction acting on the simulation environment according to the simulation environment state generated by the simulation engine and the initial model, and circularly execute for multiple times to obtain a simulation training set; using the learning engine to train an initial model according to the simulation training set to obtain an updated initial model; and sending a terminal equipment initialization request which comprises hardware resources and the updated initial model and is used for enabling the terminal equipment to configure the hardware resources and outputting an instruction acting on the real environment according to the state of the real environment. The system comprises a first scheduling device, a simulation engine, a decision engine, a learning engine and a second scheduling device. By means of the scheme, a complete decision closed loop is formed, and the purpose of conforming to expected decisions is achieved.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to a decision-making method and system based on a cloud platform. Background technique [0002] Existing decision-making systems are based on human-defined rules and cannot adapt to changes in the environment. [0003] Existing intelligent systems are only based on artificial intelligence and cannot be calibrated to the results of machine learning. Moreover, most intelligent systems focus on the learning side and cannot provide end-to-end support for decision-making. There are also a small number of end-to-end systems that are in the scientific research stage, unable to adapt to large-scale and high-availability production environments, and do not provide support for various deployment forms and learning-end or decision-making end devices. Contents of the invention [0004] In order to solve the above problems, the present invention provides a decisio...

AI Technical Summary

Problems solved by technology

[0002] Existing decision-making systems are based on human-defined rules and cannot adapt to changes in the environment

[0003] Existing intelligent systems are only based on artificial intelligence and cannot be calibrated to the results of machine learning

Moreover, most intelligent systems focus on the learning side, unable to provide end-to-end support for decision-making

There are also a small number of end-to-end systems that are in the scientific research stage, unable to adapt to large-scale and high-availability production environments, and do not provide support for various deployment forms and learning-end or decision-making end devices

Images