Image generation using adversarial attacks for imbalanced datasets

1. A method of balancing a dataset for a machine learning model, the method comprising:

receiving a machine learning model implemented in a computing system;

identifying confusing classes of few-shot classes for the machine learning model during validation;

selecting one of the confusing classes;

selecting an image from one of the few-shot classes;

computing an image perturbation such that the selected image is classified as the selected confusing class, wherein the image perturbation is computed using a gradient-ascent technique that propagates a gradient to an input image;

modifying the selected image with the computed perturbation; and

adding the modified selected image to the one few-shot class for training the machine learning model.

2. The method of claim 1 , further comprising computing a pixel update based on the gradient.

3. The method of claim 1 , wherein the selected image is modified by maximizing a posterior probability or logit of a non-true class given an input image.

4. The method of claim 1 , wherein the one confusing class is selected by:

computing a probability distribution over all classes using confusion matrix scores for a tail class;

and using the computed probability distribution to sample for a confusing class.

5. The method of claim 1 , wherein a minimum class score is computed by randomly choosing a confidence value from within 0.15 and 0.25.

6. The method of claim 1 , wherein the gradient-ascent technique is executed with a learning rate δ=0.7.

7. The method of claim 6 , further comprising stopping the gradient-ascent technique when S c′ (I′)≥s c′ or when 15 iterations is reached.

8. A computing system, comprising:

one or more processors; and

a computer-readable storage medium having computer-executable instructions stored thereupon which, when executed by the processor, cause the computing system to perform operations comprising:

selecting a confusing class of few-shot classes for a machine learning model;

selecting an image from one of the few-shot classes;

computing an image perturbation such that the selected image is classified as the selected confusing class, wherein the image perturbation is computed using a gradient-ascent technique that propagates a gradient to an input image;

modifying the selected image with the computed perturbation; and

adding the modified selected image to a batch for training the machine learning model.

9. The computing system of claim 8 , further comprising computing a pixel update based on the gradient.

10. The computing system of claim 8 , wherein the selected image is modified by maximizing a posterior probability or logit of a non-true class given an input image.

11. The computing system of claim 8 , wherein the one confusing class is selected by:

computing a probability distribution over all classes using confusion matrix scores for a tail class;

and using the computed probability distribution to sample for a confusing class.

12. A computer-readable storage medium having computer-executable instructions stored thereupon which, when executed by one or more processors of a computing device, cause the computing device to perform operations comprising:

receiving a machine learning model implemented in a computing system;

identifying confusing classes of few-shot classes for the machine learning model during validation;

selecting one of the confusing classes;

selecting an image from one of the few-shot classes;

computing an image perturbation such that the selected image is classified as the selected confusing class, wherein the image perturbation is computed using a gradient-ascent technique that propagates a gradient to an input image;

modify the selected image with the computed perturbation; and

adding the modified selected image to a batch for training the machine learning model.

13. The computer-readable storage medium of claim 12 , wherein a minimum class score is computed by randomly choosing a confidence value from within 0.15 and 0.25.

14. The computer-readable storage medium of claim 12 , wherein:

the gradient-ascent technique is executed with a learning rate δ=0.7.

15. The computer-readable storage medium of claim 14 , further comprising stopping the gradient-ascent technique when S c′ (I′)≥s c′ or when 15 iterations is reached.

16. The computer-readable storage medium of claim 12 , further comprising computing a pixel update based on the gradient.

17. The computer-readable storage medium of claim 12 , wherein the selected image is modified by maximizing a posterior probability or logit of a non-true class given an input image.