Training method, device and equipment for resource allocation system

Abstract

The invention discloses a training method, device and equipment for a resource allocation system, and the method comprises the steps: packaging a first algorithm and a first simulation engine, and obtaining a first execution program of an initial model; running the first execution program based on the initial model, and generating at least one group of situation data; and performing training operation on each group of situation data in the at least one group of situation data until execution results corresponding to the at least one group of situation data meet corresponding conditions, thereby obtaining the resource allocation system. Through the above mode, an extensible and reusable resource distribution system is established, and the resource distribution system can realize intelligent decision-making of multi-target multi-resource dynamic distribution.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a training method, device and equipment for a resource allocation system. Background technique [0002] The core of the multi-objective multi-resource allocation (MOMRA) problem is to give a correct and reliable allocation plan in a short time. However, due to the constraints of algorithm performance and the complexity of the MOMRA environment, the research on multi-objective multi-resource allocation is often limited to static and small-scale scenarios. The multi-objective multi-resource dynamic allocation (MOMRDA) problem is a repetition of the multi-objective multi-resource allocation (MOMRA) problem, and its complexity increases exponentially with the number of decisions. [0003] For the MOMRDA problem, scholars propose to evaluate the target situation through the expert system knowledge base, and then use the heuristic algorithm to calculate the amo...

AI Technical Summary

Problems solved by technology

However, the traditional expert system has disadvantages such as single logic, simple structure, fixed knowledge base content, poor reusability, etc., and the expert system is mainly used in pre-prediction activities such as reasoning and planning, and it is difficult to deal with unexpected situations

At the same time, the heuristic algorithm takes a long time to obtain the optimized allocation plan, and the scale of the problem to be dealt with is also limited, so it is difficult to meet the requirements of multi-objective

[0004] In view of the MOMRDA problem, if a distribution system is established for each environment, it will be a very heavy work, and it will also cause a lot of waste

Images