Changeflow GovPing Telecom & Technology Meta Platforms multi-task neural network patent
Routine Notice Added Final

Meta Platforms multi-task neural network patent

Favicon for changeflow.com USPTO Patent Grants - AI & Computing (G06N)
Published
Detected
Email

Summary

USPTO granted Meta Platforms Patent US12591775B1 for a multi-task neural network training process using collaborative data and matrix factorization. The patent covers systems and methods for generating interaction matrices from user-item data and training neural networks through latent vector comparisons. Patent granted March 31, 2026 to Meta Platforms, Inc.

What changed

USPTO granted Patent US12591775B1 to Meta Platforms, Inc. covering a multi-task neural network training process that generates interaction matrices from user-item collaboration data, applies matrix factorization to produce user and item vectors, and trains neural networks by comparing latent item vectors against matrix-derived item vectors. The patent names inventors Oren Sar Shalom and Yaakov Aviv and contains 17 claims under CPC classification G06N 3/08.

Technology companies and AI developers working with recommendation systems, collaborative filtering, or multi-task learning architectures should review this patent to assess potential licensing needs or freedom-to-operate implications for their own systems. Patent grants do not impose compliance deadlines but establish intellectual property rights that may affect product development strategies.

Archived snapshot

Mar 31, 2026

GovPing captured this document from the original source. If the source has since changed or been removed, this is the text as it existed at that time.

← USPTO Patent Grants

Multi-task training process based on collaborative data

Grant US12591775B1 Kind: B1 Mar 31, 2026

Assignee

Meta Platforms, Inc.

Inventors

Oren Sar Shalom, Yaakov Bibas

Abstract

Systems, apparatuses and methods provide technology that generates an interaction matrix from collaboration data, where the collaboration data represents user interactions with a plurality of items. The technology applies a matrix factorization process to the interaction matrix to generate user vectors and matrix item vectors, and executes a multi-task training process with a neural network. The multi-task training process includes processing the collaboration data with the neural network to generate latent item vectors, and training the neural network based on a comparison of the latent item vectors to the matrix item vectors.

CPC Classifications

G06N 3/08

Filing Date

2022-07-29

Application No.

17816036

Claims

17

View original document →

Get daily alerts for USPTO Patent Grants - AI & Computing (G06N)

Daily digest delivered to your inbox.

Free. Unsubscribe anytime.

About this page

What is GovPing?

Every important government, regulator, and court update from around the world. One place. Real-time. Free. Our mission

What's from the agency?

Source document text, dates, docket IDs, and authority are extracted directly from USPTO.

What's AI-generated?

The summary, classification, recommended actions, deadlines, and penalty information are AI-generated from the original text and may contain errors. Always verify against the source document.

Last updated

Classification

Agency
USPTO
Published
March 31st, 2026
Instrument
Notice
Legal weight
Non-binding
Stage
Final
Change scope
Minor
Document ID
US12591775B1

Who this affects

Applies to
Technology companies
Industry sector
5112 Software & Technology
Activity scope
Patent Grant
Geographic scope
United States US

Taxonomy

Primary area
Intellectual Property
Operational domain
Legal
Topics
Artificial Intelligence Data Privacy

Get alerts for this source

We'll email you when USPTO Patent Grants - AI & Computing (G06N) publishes new changes.

Free. Unsubscribe anytime.

You're subscribed!