Efficient causal reward auto-discovery system for edge computing

By introducing a causal discovery reward generation module into the edge computing system, the system analyzes causal relationships such as load, network status, and device heterogeneity to generate interpretable reward strategies, optimize task priority and resource allocation, solve the problem of low resource utilization efficiency in edge computing, and achieve more efficient task scheduling and more stable computing performance.

CN122111608APending Publication Date: 2026-05-29NINGBO XINGBOYUAN INTELLIGENT TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NINGBO XINGBOYUAN INTELLIGENT TECHNOLOGY CO LTD
Filing Date
2026-02-12
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing technologies in edge computing lack in-depth analysis of the inherent causal relationships of the system, resulting in low resource utilization efficiency, decreased task processing performance, and inability to meet the requirements of real-time performance and accuracy.

Method used

A reward generation module based on causal discovery is introduced. By analyzing the causal relationship between load, network status, device heterogeneity and task execution effect, a dynamic causal graph is constructed to generate an interpretable reward strategy and optimize task priority and resource allocation.

Benefits of technology

It enables precise task scheduling and efficient resource utilization in edge scenarios with limited resources and variable environments, thereby improving the system's intelligence, response speed, and reliability.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses an efficient causal reward automatic discovery system for edge computing, and belongs to the technical field of edge computing. The system comprises an edge computing box, an intelligent terminal device, an edge cloud server and a public cloud server. The core of the system is that the edge computing box is provided with a reward generation module based on causal discovery. A dynamic causal graph is constructed by analyzing historical task data, and an interpretable reward strategy is generated to optimize task priority and resource allocation. The public cloud server is provided with a global causal knowledge base for fusing causal data of each edge node, generating and issuing an enhanced global causal model and a strategy template. The combination of causal inference and automatic reward discovery enables the system to adapt to the dynamics and uncertainty of the edge environment, realizes efficient utilization of resources, precise optimization of task scheduling, and significant improvement of the overall performance and reliability of the system.
Need to check novelty before this filing date? Find Prior Art