Patent · US Active

Training strategy search using reinforcement learning

US11100643B2 · kind B2 · utility

3Cited by

0References

29Claims

0Family size

Assignee

NVIDIA Corporation · US

Inventors

Dong Yang · Somerville, US
Holger Reinhard Roth · Rockville, US
Ziyue Xu · Iowa City, US
Fausto Milletari · Munich, DE
Ling Zhang · Hefei, CN
Te-Chung Isaac Yang · Aliso Viejo, US
Daguang Xu · Princeton, US

Key dates

Filing date	Sep 11, 2019
Grant date	Aug 24, 2021
Priority date	—
Expiry date	Sep 11, 2039

Classification

Technology area (CPC G)Physics
CPC primaryG06T7/0012
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

In at least one embodiment, a reinforcement-learning-based searching approach is used to produce a training configuration for a machine-learning model. In at least one embodiment, 3D medical image segmentation is performed using learned image preprocessing parameters.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.