USPTO Grants Patent for Consensus Driven Learning

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 24th, 2026

Summary

The United States Patent and Trademark Office (USPTO) has granted a patent (US12585938B2) for a consensus driven learning system. This technology enables decentralized machine learning model training across multiple computational nodes without requiring a central server.

View original document View source feed page

What changed

The USPTO has granted patent US12585938B2, titled "Consensus driven learning," to the Government of the United States of America, as represented by the Secretary of the Navy. The patent details systems and methods for decentralized machine learning model training, allowing devices to learn from distributed datasets without a central server, high bandwidth, or robust communication architecture.

This patent grant is primarily an intellectual property matter and does not impose new regulatory obligations on regulated entities. However, companies involved in AI and machine learning development, particularly those focusing on decentralized or federated learning approaches, may find the patented technology relevant to their own intellectual property strategies or potential licensing considerations.

Source document (simplified)

← USPTO Patent Grants

Consensus driven learning

Grant US12585938B2 Kind: B2 Mar 24, 2026

Assignee

The Government of the United States of America, as represented by the Secretary of the Navy

Inventors

Kyle Crandall, Dustin Webb

Abstract

Systems and methods are provided for consensus driven learning (CDL) using machine learning (ML) to enable devices to learn a model on a data set that is distributed over several computational nodes in a decentralized manner. In an embodiment, local models are trained on local data and share model parameters in an asynchronous, decentralized, and distributed manner that imposes minimal restrictions on the topology of a communications network. Systems and methods using CDL in accordance with embodiments of the present disclosure do not require a central server to coordinate models like most conventional technologies, high bandwidth, or highly robust communication architecture between nodes.