Layered federal learning method for adaptive compressed coding

Through adaptive compression coding and edge aggregation, the problems of training asymmetry and transmission deviation between devices in hierarchical federated learning are solved, and learning efficiency and model accuracy are improved.

CN119210465AActive Publication Date: 2024-12-27HUNAN UNIV OF SCI & TECH
View PDF 3 Cites 5 Cited by

Patent Information

Application Number
CN202411579167.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-07
Publication Date
2024-12-27
Estimated Expiration
2044-11-07

AI Technical Summary

Technical Problem

In the existing stratified federated learning technology, fixed training methods lead to out-of-synchronization of training between devices, accumulation of transmission deviations, and affecting the learning effect.

Method used

Adopting hierarchical federated learning method of adaptive compression coding, by evaluating the data processing performance and network transmission performance of terminal devices, dynamically matches edge servers, performs adaptive compression coding and edge aggregation, and reduces data transmission costs and delays.

Benefits of technology

Synchronous training among devices is realized, which reduces training delay and communication overhead, and improves data transmission efficiency and model accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119210465A_ABST
    Figure CN119210465A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of electric digital data processing, and particularly discloses a hierarchical federated learning method for adaptive compression coding, which comprises the steps of hierarchical federated learning architecture construction, edge server matching, model generation performance judgment, compression ratio matching and hierarchical federated learning architecture judgment, and specifically comprises the following steps: firstly, constructing a hierarchical federated learning architecture; each layer can make full use of computing resources thereof, edge servers corresponding to the terminal devices are obtained through matching, the terminal devices are used for generating environment data training models, corresponding compression rates are obtained according to network transmission performance of the edge servers, and the compression rates are used for data compression. The edge server receives the environment data training model of each terminal device and carries out adaptive compression coding and edge aggregation in sequence according to a corresponding compression ratio, and the cloud server receives an edge aggregation result and carries out global aggregation; and finally, judging whether the cloud server to which the hierarchical federated learning architecture belongs uploads a global aggregation corresponding result to an environment resource scheduling cloud platform or not.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Hierarchical Federated Learning Method and Device for Distributed Deep Neural Networks

    CN113971090B

  • Federal learning-oriented data compression method and system

    CN118868958A

  • Federation learning method, system, terminal device and storage medium

    WO2021008017A1

Cited By

  • Layered credible federal learning method with data perception capability

    CN119808898A

  • Electromechanical equipment dynamic tracking method based on multivariate heterogeneous data and edge calculation

    CN119892587A

  • Quantum network path selection and resource allocation method and system for realizing high throughput

    CN120301823A

  • All-media all-signal integrated management platform

    CN120935094A

  • Distributed model training method and system based on training and compression layering

    CN121031719A