System and method of improving compression of predictive models

1. A computer-implemented method comprising:

generating an unlabeled simulated data set by expanding an initial data set, the initial data set including a first plurality of fact sets and the unlabeled simulated data set including a second plurality of fact sets;

generating a labeled data set by predicting the unlabeled simulated data set using a complex model to output a plurality of labels, the labeled data set including the second plurality of fact sets and the plurality of labels;

generating a plurality of intermediate predictions by predicting the second plurality of fact sets using a neural network model;

comparing the plurality of labels to the plurality of intermediate predictions to determine a measure of accuracy; and

modifying at least one of a plurality of training parameters of the neural network model based upon the measure of accuracy.

2. The computer-implemented method of claim 1 , further comprising:

training the neural network model iteratively until the measure of accuracy is within a predetermined threshold.

3. The computer-implemented method of claim 2 , further comprising:

generating a graphical depiction of the trained neural network model.

4. The computer-implemented method of claim 2 , further comprising:

receiving the trained neural network model in a computing device;

generating an unlabeled new data set based upon data collected by the computing device, the unlabeled new data set including a plurality of new fact sets; and

generating a plurality of device predictions by predicting the unlabeled new data set using the trained neural network model.

5. The computer-implemented method of claim 2 , further comprising:

sending the trained neural network model to a remote computing device to enable the remote computing device to analyze one or more unlabeled new data sets using the trained neural network model.

6. The computer-implemented method of claim 5 , wherein sending the trained neural network model to the remote computing device includes sending the trained neural network model to a mobile device of a user.

7. The computer-implemented method of claim 1 , wherein generating the unlabeled simulated data set by expanding the initial data set includes generating the unlabeled data set such that a distribution of a plurality of fact types within the second plurality of fact sets is skewed when compared to a distribution of a plurality of fact types within the first plurality of fact sets.

8. The computer-implemented method of claim 1 , further comprising:

obtaining a definition of the plurality of training parameters of the neural network model from a remote database.

9. The computer-implemented method of claim 1 , wherein generating the plurality of intermediate predictions includes:

dividing the second plurality of fact sets into fact subsets;

receiving, at each of a plurality of networked computing devices, one of the fact subsets;

generating, by each of the plurality of networked computing devices predicting a respective one of the fact subsets using the neural network model, a respective intermediate prediction; and

receiving, at a single networked computing device, the respective intermediate prediction corresponding to each of the fact subsets.

10. A computing system comprising:

one or more processors; and

one or more memories storing instructions that, when executed by the one or more processors, cause the computing system to:

generate an unlabeled simulated data set by expanding an initial data set, the initial data set including a first plurality of fact sets and the unlabeled simulated data set including a second plurality of fact sets;

generate a labeled data set by predicting the unlabeled simulated data set using a complex model to output a plurality of labels, the labeled data set including the second plurality of fact sets and the plurality of labels;

generate a plurality of intermediate predictions by predicting the second plurality of fact sets using a neural network model;

compare the plurality of labels to the plurality of intermediate predictions to determine a measure of accuracy; and

modify at least one of a plurality of training parameters of the neural network model based upon the measure of accuracy.

11. The computing system of claim 10 , wherein the instructions further comprise instructions that, when executed by the one or more processors, cause the computing system to train the neural network model iteratively until the measure of accuracy is within a predetermined threshold.

12. The computing system claim 11 , wherein the instructions further comprise instructions that, when executed by the one or more processors, cause the computing system to generate a graphical depiction of the trained neural network model.

13. The computing system of claim 11 , further comprise instructions that, when executed by the one or more processors, cause the computing system to send the trained neural network model to a remote computing device to enable the remote computing device to analyze one or more unlabeled new data sets using the trained neural network model.

14. The computing system of claim 13 , wherein the remote computing device is a mobile device of a user.

15. The computing system of claim 10 , wherein the instructions that cause the computing system to generate the unlabeled data set by expanding the initial data set further comprise instructions that cause the computing system to generate the unlabeled data set such that a distribution of a plurality of fact types within the second plurality of fact sets is skewed when compared to a distribution of a plurality of fact types within the first plurality of fact sets.

16. The computing system of claim 10 , further comprise instructions that, when executed by the one or more processors, cause the computing system to obtain a definition of the plurality of training parameters of the neural network model from a remote database.

17. A non-transitory computer readable medium storing instructions that, when executed by at least one processor of a computing device, cause the computing device to:

generate an unlabeled simulated data set by expanding an initial data set, the initial data set including a first plurality of fact sets and the unlabeled simulated data set including a second plurality of fact sets;

generate a labeled data set by predicting the unlabeled simulated data set using a complex model to output a plurality of labels, the labeled data set including the second plurality of fact sets and the plurality of labels;

generate a plurality of intermediate predictions by predicting the second plurality of fact sets using a neural network model;

compare the plurality of labels to the plurality of intermediate predictions to determine a measure of accuracy; and

modify at least one of a plurality of training parameters of the neural network model based upon the measure of accuracy.

18. The non-transitory computer readable medium of claim 17 , further comprising instructions that, when executed by the at least one processor, cause the computing device to train the neural network model iteratively until the measure of accuracy is within a predetermined threshold.

19. The non-transitory computer readable medium of claim 18 , wherein the instructions when executed by the at least one processor that cause the computing device to generate the unlabeled data set by expanding the initial data set further cause the computing device to generate the unlabeled data set such that a distribution of a plurality of fact types within the second plurality of fact sets is skewed when compared to a distribution of a plurality of fact types within the first plurality of fact sets.

20. The non-transitory computer readable medium of claim 18 , further comprising instructions that, when executed by the at least one processor, cause the computing device to generate a graphical depiction of the trained neural network model.