Analytic system for gradient boosting tree compression

A computing device comprising:

• a processor;
• and
• a non-transitory computer-readable medium operably coupled to the processor, the computer-readable medium having computer-readable instructions stored thereon that, when executed by the processor, cause the computing device to
• train a gradient boosting tree predictive model using a plurality of observation vectors, wherein each observation vector of the plurality of observation vectors includes an explanatory variable value of an explanatory variable and a response variable value for a response variable, wherein the gradient boosting tree predictive type model is trained to predict the response variable value of each observation vector based on a respective explanatory variable value of each observation vector;
• compress the trained gradient boosting tree predictive model using a compression model with a predefined penalty constant value and with a predefined array of coefficients to reduce a number of trees of the trained gradient boosting tree predictive model, wherein the compression model minimizes a sparsity norm loss function;
• and
• output the compressed, trained gradient boosting tree predictive model for predicting a new response variable value from a new observation vector.

A method of compressing a gradient boosting tree predictive model, the method comprising:

• training, by a computing device, a gradient boosting tree predictive model using a plurality of observation vectors, wherein each observation vector of the plurality of observation vectors includes an explanatory variable value of an explanatory variable and a response variable value for a response variable, wherein the gradient boosting tree predictive type model is trained to predict the response variable value of each observation vector based on a respective explanatory variable value of each observation vector;
• compressing, by the computing device, the trained gradient boosting tree predictive model using a compression model with a predefined penalty constant value and with a predefined array of coefficients to reduce a number of trees of the trained gradient boosting tree predictive model, wherein the compression model minimizes a sparsity norm loss function;
• and
• outputting, by the computing device, the compressed, trained gradient boosting tree predictive model for predicting a new response variable value from a new observation vector.