Improving the Accuracy of Least-Squares Probabilistic Classifiers

Improving the Accuracy of Least-Squares Probabilistic Classifiers
复制标题

提高最小二乘概率分类器的准确性

DOI:
10.1587/transinf.e94.d.1337
复制
发表时间:
2010
期刊:
IEICE Trans. Inf. Syst.
影响因子:
--
通讯作者:
J. Simm
J. Simm
中科院分区:
--
文献类型:
--
作者:
M. Yamada;Masashi Sugiyama;G. Wichern;J. Simm

文献摘要

参考文献

被引文献

相似文献

最小二乘概率分类器(LSPC)是核逻辑回归的一种计算效率高的替代方法。然而,为了确保其学习概率是非负的,LSPC涉及将负参数舍入为零的后处理步骤,这可能会意外地影响分类性能。为了缓解这个问题,我们提出了一个简单的替代方案,直接四舍五入分类器的负输出,而不是负参数。通过大量的实验,包括真实世界的图像分类和音频标记任务,我们证明,所提出的修改显着提高分类精度,而原来的LSPC的计算优势保持不变。
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 problem, we propose a simple alternative scheme that directly rounds up the classifier's negative outputs, not negative parameters. Through extensive experiments including real-world image classification and audio tagging tasks, we demonstrate that the proposed modification significantly improves classification accuracy, while the computational advantage of the original LSPC remains unchanged.
考虑1岁和2岁儿童的“生活质量” - 考虑托儿所的上课时间为10:00至13:00 -
DOI: --
发表时间: 2007
期刊:
影响因子: --
作者:
阿部;和子
通讯作者: 和子