Improving the Accuracy of Least-Squares Probabilistic Classifiers

Abstract

The least-squares probabilistic classifier (LSPC) is a computationally-efficient alternative to kernel logistic regression. However, to assure its learned probabilities to be non-negative, LSPC involves a post-processing step of rounding up negative parameters to zero, which can unexpectedly influence classification performance. In order to mitigate this… (More)

Topics

3 Figures and Tables

Cite this paper

@article{Yamada2011ImprovingTA, title={Improving the Accuracy of Least-Squares Probabilistic Classifiers}, author={Makoto Yamada and Masashi Sugiyama and Gordon Wichern and Jaak Simm}, journal={IEICE Transactions}, year={2011}, volume={94-D}, pages={1337-1340} }