Machine learning (ML) modeling by DNA computing

A computer program product comprising:

• a computer readable storage medium readable by two or more processors and storing instructions for execution by the two or more processors for performing a method for generating a machine learning model, comprising:
• identifying a training data set;
• defining a window for an initial beta value, wherein the initial beta value represents bias for the identified training data set in respective conditional expectations;
• regularizing the respective conditional expectations corresponding to each feature vector from the training data set, wherein two or more methods of regularization on the conditional expectations are performed in parallel for each feature vector;
• creating a plurality of ensembles of models derived from the conditional expectations resulting from the regularizing;
• and
• producing a best fit ensemble amongst the created ensembles as the machine learning model, wherein the machine learning model predicts targeted outcomes based on inputs distinctive from the training data set,
• wherein the two or more processors include one or more digital processor and one or more DNA processor.

A system comprising:

• a memory;
• two or more processors in communication with the memory;
• and
• program instructions executable by the two or more processors via the memory to perform a method for generating a machine learning model, comprising:
• identifying a training data set;
• defining a window for an initial beta value, wherein the initial beta value represents bias for the identified training data set in respective conditional expectations;
• regularizing the respective conditional expectations corresponding to each feature vector from the training data set, wherein two or more methods of regularization on the conditional expectations are performed in parallel for each feature vector;
• creating a plurality of ensembles of models derived from the conditional expectations resulting from the regularizing;
• and
• producing a best fit ensemble amongst the created ensembles as the machine learning model, wherein the machine learning model predicts targeted outcomes based on inputs distinctive from the training data set,
• wherein the two or more processors include one or more digital processor and one or more DNA processor.

A computer implemented method for generating a machine learning model, comprising:

• identifying a training data set;
• defining a window for an initial beta value, wherein the initial beta value represents bias for the identified training data set in respective conditional expectations;
• regularizing the respective conditional expectations corresponding to each feature vector from the training data set, wherein two or more methods of regularization on the conditional expectations are performed in parallel for each feature vector;
• creating a plurality of ensembles of models derived from the conditional expectations resulting from the regularizing;
• and
• producing a best fit ensemble amongst the created ensembles as the machine learning model, wherein the machine learning model predicts targeted outcomes based on inputs distinctive from the training data set.