# US 2025/0157207 A1 · Generative data augmentation with task loss guided fine-tuning

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0157207 A1
- Document ID: US20250157207A1
- Status: Published application
- Kind: A1
- Application number: 18/506,018
- Filed: 2023-11-09
- Published: 2025-05-15
- Earliest claimed: 2023-11-09
- Assignees: QUALCOMM Incorporated
- Applicants: QUALCOMM Incorporated
- Inventors: Davide ABATI; Amirhossein HABIBIAN; Auke Joris WIGGERS; Jens PETERSEN
- Web page: https://patentgrep.com/patent/US20250157207A1

## CPC

- G06N 3/045: Combinations of networks
- G06V 10/82: using neural networks
- G06N 3/0475: Generative networks
- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A method includes generating a synthetic dataset with a generative model. The method also includes tuning the generative model based on feedback from a task network that receives the synthetic dataset as input. The task network may perform image recognition. The synthetic dataset may be generated based on a set of classes and labels of the classes. The method may iteratively generate the synthetic dataset and tune the generative model, based on feedback from the task network.

## Claims

### Claim 1

An apparatus, comprising:
at least one memory storing instructions; and
at least one processor coupled to the at least one memory and configured to instruct the apparatus to:
generate a synthetic dataset with a generative model; and
tune the generative model based on feedback from a task network that receives the synthetic dataset as input.

### Claim 2 (depends on 1)

The apparatus of claim 1, in which the task network performs image recognition.

### Claim 3 (depends on 1)

The apparatus of claim 1, in which the at least one processor is further configured to instruct the apparatus to iteratively generate the synthetic dataset and tune the generative model, based on feedback from the task network.

### Claim 4 (depends on 1)

The apparatus of claim 1, in which the at least one processor is further configured to instruct the apparatus to:
obtain a prediction, with task network, based on the synthetic dataset;
calculate a loss of the task network based on the prediction from the synthetic dataset;
back propagate the loss of the task network to the generative model;
tune the generative model by reducing the loss of the task network to obtain a tuned generative model; and
generate an updated synthetic dataset with the tuned generative model.

### Claim 5 (depends on 4)

The apparatus of claim 4, in which the synthetic dataset is generated based on a set of classes and labels of the classes.

### Claim 6 (depends on 4)

The apparatus of claim 4, in which the at least one processor is further configured to instruct the apparatus to retrain the task network based on the updated synthetic dataset.

### Claim 7 (depends on 1)

The apparatus of claim 1, in which the at least one processor is further configured to instruct the apparatus to:
generate a synthetic dataset with a generative model based on a set of real images;
obtain a first prediction, with the task network, based on the synthetic dataset;
calculate a loss of the task network based on the first prediction from the synthetic dataset;
update the task network based on the loss of the network to obtain an updated task network;
calculate a loss of the updated task network based on a second prediction of the updated task network and the set of real images;
back propagate the loss of the updated task network to the generative model;
tune the generative model by reducing the loss of the updated task network to obtain a tuned generative model; and
generate an updated synthetic dataset with the tuned generative model.

### Claim 8

A method, comprising:
generating a synthetic dataset with a generative model; and
tuning the generative model based on feedback from a task network that receives the synthetic dataset as input.

### Claim 9 (depends on 8)

The method of claim 8, in which the task network performs image recognition.

### Claim 10 (depends on 8)

The method of claim 8, further comprising iteratively generating the synthetic dataset and tuning the generative model, based on feedback from the task network.

### Claim 11 (depends on 8)

The method of claim 8, further comprising:
obtaining a prediction, with task network, based on the synthetic dataset;
calculating a loss of the task network based on the prediction from the synthetic dataset;
back propagating the loss of the task network to the generative model;
tuning the generative model by reducing the loss of the task network to obtain a tuned generative model; and
generating an updated synthetic dataset with the tuned generative model.

### Claim 12 (depends on 11)

The method of claim 11, in which the synthetic dataset is generated based on a set of classes and labels of the classes.

### Claim 13 (depends on 11)

The method of claim 11, further comprising retraining the task network based on the updated synthetic dataset.

### Claim 14 (depends on 8)

The method of claim 8, further comprising:
generating a synthetic dataset with a generative model based on a set of real images;
obtaining a first prediction, with the task network, based on the synthetic dataset;
calculating a loss of the task network based on the first prediction from the synthetic dataset;
updating the task network based on the loss of the network to obtain an updated task network;
calculating a loss of the updated task network based on a second prediction of the updated task network and the set of real images;
back propagating the loss of the updated task network to the generative model;
tuning the generative model by reducing the loss of the updated task network to obtain a tuned generative model; and
generating an updated synthetic dataset with the tuned generative model.

### Claim 15

A non-transitory computer-readable medium having program code recorded thereon, the program code executed by a processor and comprising:
program code to generate a synthetic dataset with a generative model; and
program code to tune the generative model based on feedback from a task network that receives the synthetic dataset as input.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, in which the task network performs image recognition.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium of claim 15, in which the program code comprises program code to iteratively generate the synthetic dataset and tune the generative model, based on feedback from the task network.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium of claim 15, in which the program code comprises:
program code to obtain a prediction, with task network, based on the synthetic dataset;
program code to calculate a loss of the task network based on the prediction from the synthetic dataset;
program code to back propagate the loss of the task network to the generative model;
program code to tune the generative model by reducing the loss of the task network to obtain a tuned generative model; and
program code to generate an updated synthetic dataset with the tuned generative model.

### Claim 19 (depends on 18)

The non-transitory computer-readable medium of claim 18, in which the synthetic dataset is generated based on a set of classes and labels of the classes.

### Claim 20 (depends on 18)

The non-transitory computer-readable medium of claim 18, in which the program code comprises program code to retrain the task network based on the updated synthetic dataset.

### Claim 21 (depends on 15)

The non-transitory computer-readable medium of claim 15, in which the program code comprises:
program code to generate a synthetic dataset with a generative model based on a set of real images;
program code to obtain a first prediction, with the task network, based on the synthetic dataset;
program code to calculate a loss of the task network based on the first prediction from the synthetic dataset;
program code to update the task network based on the loss of the network to obtain an updated task network;
program code to calculate a loss of the updated task network based on a second prediction of the updated task network and the set of real images;
program code to back propagate the loss of the updated task network to the generative model;
program code to tune the generative model by reducing the loss of the updated task network to obtain a tuned generative model; and
program code to generate an updated synthetic dataset with the tuned generative model.

### Claim 22

An apparatus, comprising:
means for generating a synthetic dataset with a generative model; and
means for tuning the generative model based on feedback from a task network that receives the synthetic dataset as input.

### Claim 23 (depends on 22)

The apparatus of claim 22, in which the task network performs image recognition.

### Claim 24 (depends on 22)

The apparatus of claim 22, further comprising means for iteratively generating the synthetic dataset and means for tuning the generative model, based on feedback from the task network.

### Claim 25 (depends on 22)

The apparatus of claim 22, further comprising:
means for obtaining a prediction, with task network, based on the synthetic dataset;
means for calculating a loss of the task network based on the prediction from the synthetic dataset;
means for back propagating the loss of the task network to the generative model;
means for tuning the generative model by reducing the loss of the task network to obtain a tuned generative model; and
means for generating an updated synthetic dataset with the tuned generative model.

### Claim 26 (depends on 25)

The apparatus of claim 25, in which the synthetic dataset is generated based on a set of classes and labels of the classes.

### Claim 27 (depends on 25)

The apparatus of claim 25, further comprising means for retraining the task network based on the updated synthetic dataset.

### Claim 28 (depends on 22)

The apparatus of claim 22, further comprising:
means for generating a synthetic dataset with a generative model based on a set of real images;
means for obtaining a first prediction, with the task network, based on the synthetic dataset;
means for calculating a loss of the task network based on the first prediction from the synthetic dataset;
means for updating the task network based on the loss of the network to obtain an updated task network;
means for calculating a loss of the updated task network based on a second prediction of the updated task network and the set of real images;
means for back propagating the loss of the updated task network to the generative model;
means for tuning the generative model by reducing the loss of the updated task network to obtain a tuned generative model; and
means for generating an updated synthetic dataset with the tuned generative model.
