Multilabel learning with label relationships
US11769087B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jun 4, 2020 |
| Grant date | Sep 26, 2023 |
| Priority date | — |
| Expiry date | Feb 12, 2041 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N20/00
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Machine learning based method for multilabel learning with label relationships is provided. This methodology addresses the technical problem of alleviating computational complexity of training a machine learning model that generates multilabel output with constraints, especially in contexts characterized by a large volume of data, by providing a new formulation that encodes probabilistic relationships among the labels as a regularization parameter in the training objective of the underlying model. For example, the training process of the model may be configured to have two objectives. Namely, in addition to the objective of minimizing conventional multilabel loss, there is another training objective, which is to minimize penalty associated with the prediction generated by the model breaking probabilistic relationships among the labels.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.