Patents

US 12762211 B2: Region metrics for class balancing in machine learning systems

Granted
2026-10-13
Filed
2021-09-17 application 17478177
Type
Utility · 20 claims
Elsewhere
Google Patents ↗ · USPTO patent document ↗ · USPTO Patent Center record ↗

Abstract

Techniques are disclosed for an image understanding system comprising a machine learning system that applies a machine learning model to perform image understanding of each pixel of an image, the pixel labeled with a class, to determine an estimated class to which the pixel belongs. The machine learning system determines, based on the classes with which the pixels are labeled and the estimated classes, a cross entropy loss of each class. The machine learning system determines, based on one or more region metrics, a weight for each class and applies the weight to the cross entropy loss of each class to obtain a weighted cross entropy loss. The machine learning system updates the machine learning model with the weighted cross entropy loss to improve a performance metric of the machine learning model for each class.

Source: PTBLXML, file of 2026-10-13 (ipgb20261013.xml).

Assignees

AssigneeTypePlaceAlso on furp.orgSource
SRI International 676 patentsUS company or organizationMenlo Park, CASRI INTERNATIONAL Federal lobbying client EXACT high confidence: identical full name and state
SRI INTERNATIONAL Nonprofit EXACT high confidence: identical full name and state
PTBLXML, file of 2026-10-13

Inventors

InventorPlacePatentsPerson pageSource
Han-Pang ChiuWest Windsor, NJ13PTBLXML, file of 2026-10-13
Junjiao TianAtlanta, GA3PTBLXML, file of 2026-10-13
Zachary SeymourPennington, NJ7PTBLXML, file of 2026-10-13
Niluthpol C. MithunLawrenceville, NJ8PTBLXML, file of 2026-10-13

CPC categories

SubclassCategoryFirst groupSource
G06FElectric digital data processingG06F18/2415PTBLXML, file of 2026-10-13
G06NComputing arrangements based on specific computational modelsG06N20/00PTBLXML, file of 2026-10-13

Cooperative Patent Classification, as currently assigned (one row per subclass).

Assignments and security interests

No recorded assignments or security interests on file for this patent.

USPTO bulk data: PatentsView granted patents, disambiguated (PVGPATDIS) for patents, assignees, inventors, places and CPC categories; the USPTO Patent Assignment Dataset (ECORSEXC) for assignments and security interests. Patents filed in the last 20 years. No claims or full text. Links to other records are by name (and state): EXACT means the identical full name and state; FUZZY means a looser name match, with the confidence shown.