Systems for multi-task joint training of neural networks using multi-label datasets
US12243292B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 2, 2022 |
| Grant date | Mar 4, 2025 |
| Priority date | — |
| Expiry date | Aug 31, 2043 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06V40/174
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Systems and methods for multi-task joint training of a neural network including an encoder module and a multi-headed attention mechanism are provided. In one aspect, the system includes a processor configured to receive input data including a first set of labels and a second set of labels. Using the encoder module, features are extracted from the input data. Using a multi-headed attention mechanism, training loss metrics are computed. A first training loss metric is computed using the extracted features and the first set of labels, and a second training loss metric is computed using the extracted features and the second set of labels. A first mask is applied to filter the first training loss metric, and a second mask is applied to filter the second training loss metric. A final training loss metric is computed based on the filtered first and second training loss metrics.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.