Machine learning model data privacy enhancement
The solution addresses the challenge of dynamically updating ML models to reflect changes in privacy and consent by retraining ML models in response to updated information, ensuring compliance and validity of the models.
WO2025061928A1 Publication Date: 2025-03-27NOKIA TECHNOLOGIES OY
2 Cites 0 Cited by
Patent Information
- Application Number
- PCT/EP2024/076411
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-09-21
- Filing Date
- 2024-09-20
- Publication Date
- 2025-03-27
AI Technical Summary
Technical Problem
Current technologies lack mechanisms for dynamically updating machine learning (ML) models to reflect changes in privacy requirements or user consent, especially in scenarios where data is already collected and stored.
Method used
The proposed solution involves a network entity that trains an ML model and informs another network entity to update associated information. Upon receiving a retraining request indicating changes in privacy or consent, the ML model is retrained to generate a new model, ensuring data privacy modifications and anonymization.
Benefits of technology
This approach enables real-time adaptation of ML models to changing privacy and consent conditions, ensuring compliance with data protection regulations and maintaining the validity of ML models.
✦ Generated by Eureka AI based on patent content.
Abstract
Example embodiments of the present disclosure relate to machine learning (ML) model data privacy enhancement In an aspect, a network entity based on receiving a request for training a first machine learning, ML, model, train the first ML model and inform a second network entity to update information associated with the first ML model. The network entity receives, from the second network entity, a retraining request for the first ML model, the retraining request being based on the updated information and indicating a change in privacy or consent. The network entity retrains the first ML model to generate a second ML model based on the retraining request. In this way, data privacy modification and data anonymization enhancement are implemented, and the data point changes are recorded.
Need to check novelty before this filing date? Find Prior Art