Adversarial pretraining of machine learning models
US12299579B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 26, 2023 |
| Grant date | May 13, 2025 |
| Priority date | — |
| Expiry date | Sep 26, 2043 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/20081
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
This document relates to training of machine learning models. One example method involves providing a machine learning model having one or more mapping layers. The one or more mapping layers can include at least a first mapping layer configured to map components of pretraining examples into first representations in a space. The example method also includes performing a pretraining stage on the one or more mapping layers using the pretraining examples. The pretraining stage can include adding noise to the first representations of the components of the pretraining examples to obtain noise-adjusted first representations. The pretraining stage can also include performing a self-supervised learning process to pretrain the one or more mapping layers using at least the first representations of the training data items and the noise-adjusted first representations of the training data items.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.