Machine learning systems and methods for training with noisy labels
US11531852B2 · kind B2 · utility
4Cited by
13References
19Claims
0Family size
Assignee
Inventor
Key dates
| Filing date | Nov 27, 2017 |
| Grant date | Dec 20, 2022 |
| Priority date | — |
| Expiry date | Jun 11, 2040 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N7/01
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Machine learning classification models which are robust against label noise are provided. Noise may be modelled explicitly by modelling “label flips”, where incorrect binary labels are “flipped” relative to their ground truth value. Distributions of label flips may be modelled as prior and posterior distributions in a flexible architecture for machine learning systems. An arbitrary classification model may be provided within the system. The classification model is made more robust to label noise by operation of the prior and posterior distributions. Particular prior and approximating posterior distributions are disclosed.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.