arXivDaily arXiv每日学术速递 周一至周五更新
arXiv cs/0201014cs.LGcs.DS
收藏

AdaBoost重量的动态变化告诉您什么是难以分类的

The Dynamics of AdaBoost Weights Tells You What's Hard to Classify

Bruno Caprile, Cesare Furlanello, Stefano Merler

摘要:

Adaboost算法中权重的动态演化包含有关关联数据点在构建Adaboost模型中所起作用的有用信息。特别是,动态诱导的数据集分为两个(容易/困难)类的bipartition。简单点在模型的构建中没有影响力,而硬点的变化相关性可以通过与其演变相关的熵值来衡量。熵的平滑近似突出了分类最不确定的区域。当所提出的方法应用于最优采样框架时,得到了令人满意的结果。

英文摘要:

The dynamical evolution of weights in the Adaboost algorithm contains useful information about the role that the associated data points play in the built of the Adaboost model. In particular, the dynamics induces a bipartition of the data set into two (easy/hard) classes. Easy points are ininfluential in the making of the model, while the varying relevance of hard points can be gauged in terms of an entropy value associated to their evolution. Smooth approximations of entropy highlight regions where classification is most uncertain. Promising results are obtained when methods proposed are applied in the Optimal Sampling framework.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑