Article ID Journal Published Year Pages File Type
6872589 Discrete Applied Mathematics 2012 10 Pages PDF
Abstract
► When applying logical techniques for the analysis of numerical data, one first binarizes the data. We introduce the notion of robust cutpoints. ► We prove generalization error bounds that improve with increased robustness of the cutpoints. ► We discuss how these considerations can inform the algorithmic choice of cutpoints.
Keywords
Related Topics
Physical Sciences and Engineering Computer Science Computational Theory and Mathematics
Preview
Robust cutpoints in the logical analysis of numerical data
Authors
, ,