US 2022/0027757 A1 · TUNING CLASSIFICATION HYPERPARAMETERS

Abstract

In an approach to hyperparameter optimization, one or more computer processors express a hyperparameter tuning process of a model based on a type of model, one or more dimensions of a training dataset, associated loss function of the model, and associated computational constraints of the model, comprising: identifying a set of optimal hyper-rectangles based a calculated local variability and a calculated best function value; calculating a point as a representative for each identified potentially optimal hyper-rectangle by locally searching over the identified set of potentially optimal hyper-rectangles; dividing one or more hyper-rectangles in the identified set of optimal hyper-rectangles into a plurality of smaller hyper-rectangles based on each calculated point; and calculating one or more optimal hyperparameters utilizing a globally converged hyper-rectangle from the plurality of smaller hyper-rectangles. The one or more computer processors classify one or more unknown datapoints utilizing the model associated with tuned hyperparameters.

Claims20

Dependent claims

A computer-implemented method comprising:

expressing, by one or more computer processors, a hyperparameter tuning process of a model based on a type of model, one or more dimensions of a training dataset, associated loss function of the model, and associated computational constraints of the model, comprising:
identifying, by one or more computer processors, a set of optimal hyper-rectangles based a calculated local variability and a calculated best function value;
calculating, by one or more computer processors, a point as a representative for each identified potentially optimal hyper-rectangle by locally searching over the identified set of potentially optimal hyper-rectangles;
dividing, by one or more computer processors, one or more hyper-rectangles in the identified set of optimal hyper-rectangles into a plurality of smaller hyper-rectangles based on each calculated point;
calculating, by one or more computer processors, one or more optimal hyperparameters utilizing a globally converged hyper-rectangle from the plurality of smaller hyper-rectangles;
and
classifying, by one or more computer processors, one or more unknown datapoints utilizing the model associated with tuned hyperparameters.

4 more elements

A computer program product comprising:

one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the stored program instructions comprising:
program instructions to express a hyperparameter tuning process of a model based on a type of model, one or more dimensions of a training dataset, associated loss function of the model, and associated computational constraints of the model, comprise:
program instructions to identify a set of optimal hyper-rectangles based a calculated local variability and a calculated best function value;
program instructions to calculate a point as a representative for each identified potentially optimal hyper-rectangle by locally searching over the identified set of potentially optimal hyper-rectangles;
program instructions to divide one or more hyper-rectangles in the identified set of optimal hyper-rectangles into a plurality of smaller hyper-rectangles based on each calculated point;
program instructions to calculate one or more optimal hyperparameters utilizing a globally converged hyper-rectangle from the plurality of smaller hyper-rectangles;
and
program instructions to classify one or more unknown datapoints utilizing the model associated with tuned hyperparameters.

5 more elements

A computer system comprising:

one or more computer processors;
one or more computer readable storage media;
and
program instructions stored on the computer readable storage media for execution by at least one of the one or more processors, the stored program instructions comprising:
program instructions to express a hyperparameter tuning process of a model based on a type of model, one or more dimensions of a training dataset, associated loss function of the model, and associated computational constraints of the model, comprise:
program instructions to identify a set of optimal hyper-rectangles based a calculated local variability and a calculated best function value;
program instructions to calculate a point as a representative for each identified potentially optimal hyper-rectangle by locally searching over the identified set of potentially optimal hyper-rectangles;
program instructions to divide one or more hyper-rectangles in the identified set of optimal hyper-rectangles into a plurality of smaller hyper-rectangles based on each calculated point;
program instructions to calculate one or more optimal hyperparameters utilizing a globally converged hyper-rectangle from the plurality of smaller hyper-rectangles;
and
program instructions to classify one or more unknown datapoints utilizing the model associated with tuned hyperparameters.

8 more elements

Dependencies are inferred from the claim text.

Tuning classification hyperparameters

Life of this patent

Tuning classification hyperparameters

Life of this patent

Abstract

Claims20