Back to results
US 2022/0188639 A1Published application
Semi-supervised learning of training gradients via task generation
AssigneeInternational Business Machines Corporation
InventorsLeonid Karlinsky, Joseph Shtok
Life of this patent
Estimating the term…
Abstract
In an approach for augmenting a neural network with a self-supervised mechanism, a processor trains a first neural network using labeled data, the first neural network configured for a main task. A processor trains a second neural network using the labeled data and unlabeled data, the second neural network being an additional component to the first neural network. A processor computes a gradient using a second loss of the second neural network based on the unlabeled data.
Claims20
A computer-implemented method comprising:
- training, by one or more processors, a first neural network using labeled data, the first neural network configured for a main task;
- training, by one or more processors, a second neural network using the labeled data and unlabeled data, the second neural network being an additional component to the first neural network;
- and
- computing, by one or more processors, a gradient using a second loss of the second neural network based on the unlabeled data.
1 more element
A computer program product comprising:
- one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising:
- program instructions to train a first neural network using labeled data, the first neural network configured for a main task;
- program instructions to train a second neural network using the labeled data and unlabeled data, the second neural network being an additional component to the first neural network;
- and
- program instructions to compute a gradient using a second loss of the second neural network based on the unlabeled data.
2 more elements
A computer system comprising:
- one or more computer processors, one or more computer readable storage media, and program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the program instructions comprising:
- program instructions to train a first neural network using labeled data, the first neural network configured for a main task;
- program instructions to train a second neural network using the labeled data and unlabeled data, the second neural network being an additional component to the first neural network;
- and
- program instructions to compute a gradient using a second loss of the second neural network based on the unlabeled data.
2 more elements
Dependencies are inferred from the claim text.