Systems and methods for generating federated learning models

By clustering and validating user data profiles across client devices, the system generates accurate federated learning models that are representative of the entire user data population, addressing the issue of skewed data and improving model precision in federated learning systems.

WO2025006510A1PCT designated stage expired Publication Date: 2025-01-02CAPITAL ONE SERVICES LLC
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
PCT/US2024/035485
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-06-26
Filing Date
2024-06-25
Publication Date
2025-01-02

AI Technical Summary

Technical Problem

Current federated learning systems fail to accurately account for skewed user data at client devices, leading to machine learning models that lack accuracy and precision, as they do not represent the user data across all client devices effectively.

Method used

The system generates anonymized user data profiles at local devices, clusters them to ensure representation, and trains machine learning models on each cluster, aggregating weights to create a combined model that is validated across other clusters for prediction accuracy, thereby generating a model that is representative of the entire user data population.

Benefits of technology

This approach results in more accurate and representative federated learning models by ensuring that skewed user data does not impact the centralized model generation, improving the overall prediction accuracy and model performance across all client devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2024035485_02012025_PF_FP_ABST
    Figure US2024035485_02012025_PF_FP_ABST
Patent Text Reader

Abstract

Methods and systems for generating federated learning models. In some aspects, the system receives, from each client device, user data profiles that are anonymized with respect to users associated with user data stored locally at a client device. The system processes the user data profiles to generate a plurality of clusters. For each cluster, the system transmits, to one or more client devices corresponding to a cluster, a first instruction to train a machine learning model on user data corresponding to user data profiles included in the cluster and a second instruction to validate the machine learning model with respect to user data corresponding to one or more clusters of the plurality of clusters other than the cluster to generate a prediction accuracy metric. The system determines, from the plurality of clusters, a first cluster based on associated prediction accuracy metrics.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Machine learning heterogeneous edge device, method, and system

    US20180253665A1

  • Integrated noise generation for adversarial training

    US20210117771A1

  • Federated machine learning using locality sensitive hashing

    US20220076169A1

  • Machine learning system and method, integration server, information processing apparatus, program, and inference model creation method

    US20220172844A1

  • Framework for machine-learning model segmentation

    US20220405650A1