Method and apparatus for improved regression modeling
AssigneeInternational Business Machines Corporation
InventorsGerald J. Tesauro, Kilian Q. Weinberger
Life of this patent
Estimating the term…
Abstract
The present invention is a method and an apparatus for improved regression modeling to address the curse of dimensionality, for example for use in data analysis tasks. In one embodiment, a method for analyzing data includes receiving a set of exemplars, where at least two of the exemplars include an input pattern (i.e., a point in an input space) and at least one of the exemplars includes a target value associated with the input pattern. A function approximator and a distance metric are then initialized, where the distance metric computes a distance between points in the input space, and the distance metric is adjusted such that an accuracy measure of the function approximator on the set of exemplars is improved.
Claims35
A method for generating a repression model, given a set of one or more exemplars, the method using a processor to perform steps comprising:
- receiving the set of one or more exemplars, where at least two of the one or more exemplars comprise an input pattern and at least one of the one or more exemplars comprises a target value associated with the input pattern;
- initializing a distance metric, where the distance metric computes a distance between pairs of input patterns;
- initializing a function approximator;
- adjusting the distance metric such that an accuracy measure of the function approximator on the set of one or more exemplars is improved;
- and
- outputting at least one of the adjusted distance metric and the function approximator to an application.
A computer readable storage medium containing an executable program for generating a regression model, given a set of one or more exemplars, where the program performs the steps of:
- receiving the set of one or more exemplars, where at least two of the one or more exemplars comprise an input pattern and at least one of the one or more exemplars comprises a target value associated with the input pattern;
- initializing a distance metric, where the distance metric computes a distance between pairs of input patterns;
- initializing a function approximator;
- adjusting the distance metric such that an accuracy measure of the function approximator on the set of one or more exemplars is improved;
- and
- outputting at least one of the adjusted distance metric and the function approximator to an application.
A system for data generating a regression model, given a set of one or more exemplars, comprising:
- one or more input/output devices configured for receiving the set of one or more exemplars, where at least two of the one or more exemplars comprise an input pattern and at least one of the one or more exemplars comprises a target value associated with the input pattern;
- and
- a regression modeling module configured for:
- initializing a distance metric, where the distance metric computes a distance between pairs of input patterns;
- initializing a function approximator;
- and
- adjusting the distance metric such that an accuracy measure of the function approximator on the set of one or more exemplars is improved,
- wherein the one or more input/output devices are further configured for outputting at least one of the adjusted distance metric and the function approximator to an application.
Dependencies are inferred from the claim text.