Samsung Patent for Optimal Neural Network Model Generation

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 25th, 2026

Summary

The USPTO has granted Samsung Electronics a patent for a method to generate optimal neural network models. The patent describes a process for identifying and retaining the most effective layers of a neural network to improve efficiency and accuracy.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has granted Samsung Electronics a patent (US12585951B2) for an innovative method and electronic device for generating optimal neural network (NN) models. The patented technology focuses on optimizing NN architecture by determining intermediate outputs and accuracy scores at various exit gates, identifying the earliest exit gate that produces an output close to the final output, and subsequently removing less effective layers and gates. This method aims to create more efficient and accurate NN models.

This patent grant is primarily of interest to technology companies, particularly those involved in AI and machine learning development. While it does not impose new regulatory obligations or compliance deadlines on regulated entities, it represents a significant development in AI technology. Compliance officers in the technology sector should be aware of this patent as it pertains to intellectual property in AI model generation, potentially impacting competitive landscapes and future technological advancements in the field.

Source document (simplified)

← USPTO Patent Grants

Method and electronic device for generating optimal neural network (NN) model

Grant US12585951B2 Kind: B2 Mar 24, 2026

Assignee

SAMSUNG ELECTRONICS CO., LTD.

Inventors

Mayukh Das, Brijraj Singh, Pradeep Nelahonne Shivamurthappa, Aakash Kapoor, Rajath Elias Soans, Soham Vijay Dixit, Sharan Kumar Allur, Venkappa Mala

Abstract

A method for generating an optimal neural network (NN) model may include determining intermediate outputs of the NN model by passing an input dataset through each intermediate exit gate of the plurality of intermediate exit gates, determining an accuracy score for each intermediate exit gate of the plurality of intermediate exit gates based on a comparison of the final output of the NN model with the intermediate output, identifying an earliest intermediate exit gate that produces the intermediate output closer to the final output based on the accuracy score, and generating the optimal NN model by removing remaining layers of the plurality of layers and remaining intermediate exit gates of the plurality of intermediate exit gates located after the determined earliest intermediate exit gate.