/Patent Tracker
GoogleGranted patentAI & softwareInterest 4/10

Google patents method to train AI models with private data

Patent drawing
Patent drawing from the official filing.

Google has been granted a patent for a technique that helps train machine learning models while keeping the data labels private. The method adds noise to the calculations to protect privacy and then uses a mathematical step called projection to clean up that noise, allowing the model to learn effectively even when the data is heavily protected.

Why it mattersThis is interesting for tech enthusiasts because it offers a specific way to balance the need for accurate AI models with the growing demand for data privacy.

In the filing’s own words

Aspects of the disclosure are directed to implementing a projection-based stochastic gradient descent technique that maintains label differential privacy when training one or more machine learning models. The technique includes denoising gradients by exploiting projections when training the machine learning models to improve performance of the trained machine learning models while maintaining label differential privacy. For instance, the projection-based stochastic gradient descent technique can improve performance of machine learning models in higher-privacy regimes, such as digital content management.

What exactly they claim (claim 1)

1. A method for training a machine learning model with differentially private labels comprising: receiving, by one or more processors, a training dataset and plurality of model weights; computing, by the one or more processors, a plurality of gradients from the training dataset and plurality of model weights based on one or more parameters for training the machine learning model; aggregating, by the one or more processors, the plurality of gradients and adding, by the one or more processors, noise to the plurality of gradients based on a privacy parameter to generate a noisy aggregated gradient; denoising, by the one or more processors, the noisy aggregated gradient via projecting to generate a projection-based gradient; and updating, by the one or more processors, the plurality of model weights to generate a plurality of updated model weights based on the projection-based gradient.

Facts

Company
Google (Google LLC)
Official title
Learning with label differential privacy via projections
Patent number
US12755205B2
Application number
18530381
Filed
Dec 6, 2023
Granted
Oct 6, 2026
Inventors
6
Claims
20

Classified under

G06N 20/00 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.