Patent · US Active

Systems and methods for martingale boosting in machine learning

US8036996B2 · kind B2 · utility

8Cited by

3References

15Claims

0Family size

Assignee

The Trustees of Columbia University in the City of New York · US

Inventors

Philip M. Long · Palo Alto, US
Rocco A. Servedio · New York, US
Roger N. Anderson · New York, US
Albert Boulanger · New York, US

Key dates

Filing date	Mar 10, 2008
Grant date	Oct 11, 2011
Priority date	—
Expiry date	Apr 4, 2029

Classification

Technology area (CPC G)Physics
CPC primaryG06N20/20
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Boosting algorithms are provided for accelerated machine learning in the presence of misclassification noise. In an exemplary embodiment, a machine learning method having multiple learning stages is provided. Each learning stage may include partitioning examples into bins, choosing a base classifier for each bin, and assigning an example to a bin by counting the number of positive predictions previously made by the base classifier associated with the bin.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.