Technique for identifying features

1. An electronic device, comprising:

one or more processing circuits;

memory configured to store program instructions, wherein, when executed by the one or more processing circuits, the program instructions cause the electronic device to perform one or more operations comprising:

calculating combinations of feature vectors and noise vectors, wherein a given combination in the combinations corresponds to a given feature vector in the feature vectors and a given noise vector in the noise vectors, wherein a number of the feature vectors exceeds one hundred and a number of the combinations exceeds ten thousand, and wherein the number of feature vectors is at least an order of magnitude larger than a number of entries in the given feature vector;

determining statistical associations between types of events and the combinations, wherein a given statistical association in the statistical associations corresponds to the types of events and the given combination;

computing a noise threshold associated with a group of combinations in the combinations, wherein the noise threshold corresponds to a first aggregate property of the group of combinations, wherein the first aggregate property comprises numbers of occurrences of the feature vectors in the group of combinations, and wherein the group of combinations have statistical associations equal to or greater than the noise threshold;

selecting a subset of the feature vectors based at least in part on the first aggregate property of the group of combinations; and

training a supervised machine-learning model using the types of events and a training dataset that is based at least in part on the selected subset of the feature vectors, wherein training the supervised machine-learning model comprises determining parameters in the supervised machine-learning model and identifying feature vectors in the subset of feature vectors to use in the supervised machine-learning model, and wherein the supervised machine-learning model predicts the types of events based at least in part on the identified feature vectors and the parameters.

2. The electronic device of claim 1 , wherein the combinations are determined based at least in part on mathematical operations; and

wherein the given combination is based at least in part on a given mathematical operation in the mathematical operations.

3. The electronic device of claim 2 , wherein the subset of the feature vectors is selected based at least in part on a second aggregate property; and

wherein the second aggregate property comprises numbers of occurrences of the mathematical operations for the feature vectors in the group of combinations.

4. The electronic device of claim 1 , wherein the feature vectors comprise genetic feature vectors.

5. The electronic device of claim 4 , wherein the genetic feature vectors comprise feature vectors associated with deoxyribonucleic acid.

6. The electronic device of claim 1 , wherein the types of events comprise instances of occurrence of an event and instances of absence of the event.

7. The electronic device of claim 1 , wherein the types of events comprise responses to a first type of treatment.

8. The electronic device of claim 1 , wherein the noise threshold is computed based at least in part on at least one of: stability of rankings associated with at least a pair of subsets of the combinations having statistical associations equal to or greater than the noise threshold, in which a given ranking in the rankings is based at least in part on a second aggregate property of the given subset of the combinations in the subsets of the combinations; or differences between autocorrelations and cross-correlations of the combinations having the statistical associations equal to or greater than the noise threshold.

9. The electronic device of claim 8 , wherein the second aggregate property is different from the first aggregate property.

10. The electronic device of claim 1 , wherein the feature vectors are associated with one of: an individual, or a group of individuals.

11. The electronic device of claim 1 , wherein the noise vectors comprise random or pseudorandom numbers having mean amplitudes corresponding to a statistical characteristic of the feature vectors.

12. The electronic device of claim 1 , wherein a mean of frequencies of occurrence of categorical values in the noise vectors approximately match a mean of frequencies of occurrence of the categorical values in the feature vectors.

13. The electronic device of claim 1 , wherein the one or more operations comprise generating a predictive model based at least in part on the subset of feature vectors, the types of events and a supervised-learning technique; and

wherein the predictive model provides a recommendation or a prediction as an output from the predictive model based at least in part on values for the subset of the feature vectors as inputs to the predictive model.

14. A non-transitory computer-readable storage medium for use in conjunction with an electronic device, the computer-readable storage medium storing program instructions that, when executed by the electronic device, causes the electronic device to perform one or more operations comprising:

calculating combinations of feature vectors and noise vectors, wherein a given combination in the combinations corresponds to a given feature vector in the feature vectors and a given noise vector in the noise vectors, wherein a number of the feature vectors exceeds one hundred and a number of the combinations exceeds ten thousand;

determining statistical associations between types of events and the combinations, wherein a given statistical association in the statistical associations corresponds to the types of events and the given combination;

computing a noise threshold associated with a group of combinations in the combinations, wherein the noise threshold corresponds to a first aggregate property of the group of combinations, wherein the first aggregate property comprises numbers of occurrences of the feature vectors in the group of combinations, and wherein the group of combinations have statistical associations equal to or greater than the noise threshold;

selecting a subset of the feature vectors based at least in part on the first aggregate property of the group of combinations; and

training a supervised machine-learning model using the types of events and a training dataset that is based at least in part on the selected subset of the feature vectors, wherein training the supervised machine-learning model comprises determining parameters in the supervised machine-learning model and identifying feature vectors in the subset of feature vectors to use in the supervised machine-learning model, and wherein the supervised machine-learning model predicts the types of events based at least in part on the identified feature vectors and the parameters.

15. The computer-readable storage medium of claim 14 , wherein the combinations are determined based at least in part on mathematical operations; and

wherein the given combination is based at least in part on a given mathematical operation in the mathematical operations.

16. The computer-readable storage medium of claim 15 , wherein the subset of the feature vectors is selected based at least in part on a second aggregate property; and

wherein the second aggregate property comprises numbers of occurrences of the mathematical operations for the feature vectors in the group of combinations.

17. The computer-readable storage medium of claim 14 , wherein the noise threshold is computed based at least in part on at least one of: stability of rankings associated with at least a pair of subsets of the combinations having statistical associations equal to or greater than the noise threshold, in which a given ranking in the rankings is based at least in part on a second aggregate property of the given subset of the combinations in the subsets of the combinations; or differences between autocorrelations and cross-correlations of the combinations having the statistical associations equal to or greater than the noise threshold.

18. The computer-readable storage medium of claim 17 , wherein the second aggregate property is different from the first aggregate property.

19. The computer-readable storage medium of claim 14 , wherein the noise vectors comprise random or pseudorandom numbers having mean amplitudes corresponding to a statistical characteristic of the feature vectors.

20. A method for selecting a subset of feature vectors, comprising:

by an electronic device:

calculating combinations of feature vectors and noise vectors, wherein a given combination in the combinations corresponds to a given feature vector in the feature vectors and a given noise vector in the noise vectors, wherein a number of the feature vectors exceeds one hundred and a number of the combinations exceeds ten thousand;

determining statistical associations between types of events and the combinations, wherein a given statistical association in the statistical associations corresponds to the types of events and the given combination;

computing a noise threshold associated with a group of combinations in the combinations, wherein the noise threshold corresponds to a first aggregate property of the group of combinations, wherein the first aggregate property comprises numbers of occurrences of the feature vectors in the group of combinations, and wherein the group of combinations have statistical associations equal to or greater than the noise threshold;

selecting a subset of the feature vectors based at least in part on the first aggregate property of the group of combinations; and

training a supervised machine-learning model using the types of events and a training dataset that is based at least in part on the selected subset of the feature vectors, wherein training the supervised machine-learning model comprises determining parameters in the supervised machine-learning model and identifying feature vectors in the subset of feature vectors to use in the supervised machine-learning model, and wherein the supervised machine-learning model predicts the types of events based at least in part on the identified feature vectors and the parameters.