| CPC G06F 11/3696 (2013.01) [G06F 11/3664 (2013.01); G06F 16/2425 (2019.01); G06N 20/20 (2019.01)] | 12 Claims |
|
1. A computer-implemented method for generating one or more outputs, comprising:
providing code in a high-level language to a processor, the code comprising one or more statements defining one or more properties of a desired predictive output;
determining by the processor that one or more properties of the desired predictive output are undefined in the code;
defining by the processor at least one of the one or more undefined properties using a machine learning algorithm;
incorporating by the processor the at least one property defined using the machine learning algorithm into the code in at least one statement to provide a first modified version of the code;
generating by the processor a resulting output based at least partly on the first modified version of the code; and
redefining by the processor at least one property of the resulting output defined using the machine learning algorithm to generate a redefined resulting output,
wherein at least one property defined using the machine learning algorithm is redefined automatically by the processor based on an associated level of performance of one or more previous resulting outputs; and
generating by the processor the redefined resulting output.
|