Combined Deep Learning Inference and Compression Using Sensed Data

ChangeBridge: Patent Apps - AI & Computing (G06N)

Published March 26th, 2026

Detected March 31st, 2026

Summary

USPTO published patent application US20260087312A1 for a device and method combining deep learning inference with data compression using sensed data. The system encodes sensed data locally, transmits it in batches to a remote computing system, and receives optimized encoder and prediction models in return. Inventors: Damian Kelly, Megan O'Brien, Gregory Buckley, Colleen B. Caveney.

View original document View source feed page

What changed

USPTO published a patent application for a system that performs deep learning inference and compression on edge devices. The device encodes first sensed data using a first encoder, predicts behavior using a first prediction model, and stores the encoded data. The device transmits the encoded data in a batch to a remote computing system, which returns a refined second encoder and second prediction model based on the transmitted data. The application covers CPC classifications G06N 3/045 and G06N 3/088 (neural networks and deep learning architectures).

This is a routine patent application publication. Technology companies developing edge AI devices, IoT systems, or autonomous systems may wish to review the claims for potential freedom-to-operate concerns. The application (No. 19409228) was filed December 4, 2025, and published March 26, 2026. No compliance actions are required. Patent prosecution typically takes 2-3 years before grant or abandonment.

Source document (simplified)

← USPTO Patent Applications

COMBINED DEEP LEARNING INFERENCE AND COMPRESSION USING SENSED DATA

Application US20260087312A1 Kind: A1 Mar 26, 2026

Inventors

Damian Kelly, Megan O’Brien, Gregory Buckley, Colleen B. Caveney

Abstract

An example device is configured to encode first sensed data using a first encoder and to predict a first behavior based on the encoded first sensed data to create a first prediction using a first prediction model. The example device is configured to store the encoded first sensed data in the one or more memory units. The example device is configured to control the communication unit to transmit the encoded first sensed data in a first batch to a computing system. The example device is configured to receive, from the computing system via the communication unit, a second encoder, the second encoder being based at least in part on the encoded first sensed data. The example device is also configured to receive, from the computing system via the communication unit, a second prediction model, the second prediction model being based at least in part on the encoded first sensed data.