arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 0809.4632cs.LG
收藏

代理学习--一种半监督分类方法

Surrogate Learning - An Approach for Semi-Supervised Classification

Thomson Reuters Research and Development(汤森路透研发部)

Sriharsha Veeramachaneni, Ravikumar Kondadadi

摘要:

我们认为学习分类器的任务,从特征空间$\mathcal{X}$的类集$\mathcal{Y} = \{0, 1\}$,当功能可以划分成类条件独立的功能集$\mathcal{X}_1$和$\mathcal{X}_2$。我们展示了一个令人惊讶的事实,即类条件独立性可以用来表示原始的学习任务:1)学习从$\mathcal{X}_2$到$\mathcal{X}_1$的分类器; 2)学习特征集$\mathcal{X}_1$的类条件分布。这一事实可以用于半监督学习,因为前一个任务可以完全从未标记的样本中完成。我们目前的实验评估的想法在两个现实世界中的应用。

英文摘要:

We consider the task of learning a classifier from the feature space $\mathcal{X}$ to the set of classes $\mathcal{Y} = \{0, 1\}$, when the features can be partitioned into class-conditionally independent feature sets $\mathcal{X}_1$ and $\mathcal{X}_2$. We show the surprising fact that the class-conditional independence can be used to represent the original learning task in terms of 1) learning a classifier from $\mathcal{X}_2$ to $\mathcal{X}_1$ and 2) learning the class-conditional distribution of the feature set $\mathcal{X}_1$. This fact can be exploited for semi-supervised learning because the former task can be accomplished purely from unlabeled samples. We present experimental evaluation of the idea in two real world applications.

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。

↑