OpenTechnical
This is a record of a document published elsewhere, first published by the United States Patent and Trademark Office, as US-2026097491-A1, on 9 April 2026. OpenTechnical did not first disclose it. The identifier, hashes and timestamps below record what this library served and when, not when the material became public. View the original.

OT-2026-000000946

Training policy neural networks in simulation using scene synthesis machine learning models

Published
Permanent URL
Classification
Robotics & Automation › Robot Architectures
Inventors
BRUSSEE ARTHUR KARL, BYRAVAN ARUNKUMAR, HASENCLEVER LEONARD, HUMPLIK JAN, NORI FRANCESCO
Assignee
DEEPMIND TECH LIMITED
Original source
US-2026097491-A1, the United States Patent and Trademark Office, 9 April 2026
Document SHA-256
cf4395693577f58a94798a53fea3806eb53f7a93eeaf4fd77e83521191b92f6d

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a policy neural network for use in controlling a robot. In particular, the policy neural network can be trained in simulation using images generated by a scene synthesis machine learning model.

Publication information

Plain textMarkdownManifestIntegrity record

Citation

OpenTechnical Publication OT-2026-000000946, “Training policy neural networks in simulation using scene synthesis machine learning models,” published September 15, 2026 at 19:03:37 UTC, https://opentechnical.org/p/OT-2026-000000946.

To cite specific content, add the paragraph or figure numbers.

Terms

This record reproduces a document published elsewhere and is provided for reference and study. Rights in the original remain with its owner. See the publication policy.