Projected vector modification as mitigation for machine learning model string stuffing

1. A computer-implemented method comprising:

receiving an artifact;

extracting features from the artifact and populating a vector;

reducing features in the vector using a feature reduction operation to result in a modified vector having a plurality of buckets;

identifying a presence of predetermined types of feature within buckets of the modified vector influencing a score above a pre-determined threshold;

attenuating a contribution of the identified features within the high influence buckets of the modified vector by reducing a contribution of specific string features in the high influence buckets while retaining a contribution of non-string features in the high influence buckets without modification;

inputting the modified vector into a classification model to generate a score; and

providing the score to a consuming application or process.

2. The method of claim 1 , wherein the classification model characterizes the artifact as being malicious or benign to access, execute, or continue to execute.

3. The method of claim 2 further comprising: preventing access or execution of the artifact when the classification model characterizes the artifact as being malicious.

4. The method of claim 1 , wherein the classification model is a machine learning model trained using a training data set and providing a continuous scale output.

5. The method of claim 1 , wherein the machine learning model comprises one or more of: a logistic regression model, a neural network, a convolutional neural network, a recurrent neural network, a generative adversarial network, a support vector machine, a random forest, or a Bayesian model.

6. The method of claim 1 , wherein the predetermined types of features comprises alphanumeric strings.

7. The method of claim 1 further comprising:

inputting a plurality of vectorized malware samples into the classification model;

obtaining a plurality of scores based on the inputted vectorized malware samples; and

identifying, based on the classifications, the predetermined types of features within buckets of the vectorized malware samples that influence the scores above the pre-determined threshold.

8. The method of claim 1 , wherein the feature reduction operation comprises one or more of: principal component analysis (PCA), random projection matrices (RPM), linear discriminant analysis, canonical correlation analysis, singular value decomposition (SVD), latent semantic indexing (LSI), discrete cosine transform (DCT), randomized principal component analysis (RPCA), or large sample randomized principal component analysis (LS-RPCA).

9. A system comprising:

at least one data processor; and

memory including instructions which, when executed by the at least one data processor, result in operations comprising:

receiving an artifact;

extracting features from the artifact and populating a vector;

reducing features in the vector using a feature reduction operation to result in a modified vector having a plurality of buckets;

identifying a presence of predetermined types of features within buckets of the modified vector influencing a score above a pre-determined threshold by reducing a contribution of specific string features in the high influence buckets while retaining a contribution of non-string features in the high influence buckets without modification;

attenuating a contribution of the identified features within the high influence buckets of the modified vector;

inputting the modified vector into a classification model to generate a score; and

providing the score to a consuming application or process.

10. The system of claim 9 , wherein the classification model characterizes the artifact as being malicious or benign to access, execute, or continue to execute.

11. The system of claim 10 , wherein the operations further comprise: preventing access or execution of the artifact when the classification model characterizes the artifact as being malicious.

12. The system of claim 9 , wherein the classification model is a machine learning model trained using a training data set and providing a continuous scale output.

13. The system of claim 9 , wherein the machine learning model comprises one or more of: a logistic regression model, a neural network, a convolutional neural network, a recurrent neural network, a generative adversarial network, a support vector machine, a random forest, or a Bayesian model.

14. The system of claim 9 , wherein the predetermined types of features comprises alphanumeric strings.

15. The system of claim 9 , wherein the operations further comprise:

inputting a plurality of vectorized malware samples into the classification model;

obtaining a plurality of scores based on the inputted vectorized malware samples; and

identifying, based on the classifications, the predetermined types of features within buckets of the vectorized malware samples that influence the scores above the pre-determined threshold.

16. The system of claim 9 , wherein the feature reduction operation comprises one or more of: principal component analysis (PCA), random projection matrices (RPM), linear discriminant analysis, canonical correlation analysis, singular value decomposition (SVD), latent semantic indexing (LSI), discrete cosine transform (DCT), randomized principal component analysis (RPCA), or large sample randomized principal component analysis (LS-RPCA).

17. A non-transitory computer program product including instructions which, when executed by at least one computing device, result in operations comprising: receiving an artifact;

extracting features from the artifact and populating a vector;

reducing features in the vector using a feature reduction operation to result in a modified vector having a plurality of buckets;

identifying a presence of predetermined types of features within buckets of the modified vector influencing a score above a pre-determined threshold;

attenuating a contribution of the identified features within the high influence buckets of the modified vector by reducing a contribution of specific string features in the high influence buckets while retaining a contribution of non-string features in the high influence buckets without modification;

inputting the modified vector into a classification model to generate a score; and

providing the score to a consuming application or process.

18. The computer program product of claim 17 , wherein the classification model characterizes the artifact as being malicious or benign to access, execute, or continue to execute, and wherein the operations further comprise: preventing access or execution of the artifact when the classification model characterizes the artifact as being malicious.

19. The computer program product of claim 18 , wherein the classification model is a machine learning model trained using a training data set and providing a continuous scale output, and wherein the machine learning model comprises one or more of: a logistic regression model, a neural network, a convolutional neural network, a recurrent neural network, a generative adversarial network, a support vector machine, a random forest, or a Bayesian model.

20. The computer program product of claim 19 , wherein the operations further comprise:

inputting a plurality of vectorized malware samples into the classification model;

obtaining a plurality of scores based on the inputted vectorized malware samples; and

identifying, based on the classifications, the predetermined types of features within buckets of the vectorized malware samples that influence the scores above the pre-determined threshold.