arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1612.02802cs.LGcs.HC
收藏

小样本预测中特征相似性的交互式先验提取

Interactive Prior Elicitation of Feature Similarities for Small Sample Size Prediction

Helsinki Institute for Information Technology HIIT(赫尔辛基信息技术研究所);Aalto University(阿尔托大学)

Homayun Afrabandpey, Tomi Peltola, Samuel Kaski

摘要:

在“小$n$,大$p$“条件下的回归,即小样本量$n$和学习数据集中的大量特征$p$,是一种循环设置,其中从数据学习是困难的。利用关于特征的关系的先验知识,可以有效地减少$p$,但是对于专家来说,解释这种先验知识是困难的。在本文中,我们介绍了一种新的方法,用于引发专家先验知识的相似性的预测任务中的功能的作用。其关键思想是使用交互式多维标度(MDS)类型的散点图显示的功能,以引出的相似性关系,然后使用的预测参数的先验分布中引出的关系。具体而言,为了学习用贝叶斯线性回归预测目标变量,使用特征关系来构建具有回归系数的全协方差矩阵的高斯先验。在模拟和真实用户对文本数据的实验中,我们的方法的评估证实了特征相似性的先验启发提高了预测精度。此外,启发与交互式散点图显示优于直接启发,其中用户选择功能对从功能列表。

英文摘要:

Regression under the "small $n$, large $p$" conditions, of small sample size $n$ and large number of features $p$ in the learning data set, is a recurring setting in which learning from data is difficult. With prior knowledge about relationships of the features, $p$ can effectively be reduced, but explicating such prior knowledge is difficult for experts. In this paper we introduce a new method for eliciting expert prior knowledge about the similarity of the roles of features in the prediction task. The key idea is to use an interactive multidimensional-scaling (MDS) type scatterplot display of the features to elicit the similarity relationships, and then use the elicited relationships in the prior distribution of prediction parameters. Specifically, for learning to predict a target variable with Bayesian linear regression, the feature relationships are used to construct a Gaussian prior with a full covariance matrix for the regression coefficients. Evaluation of our method in experiments with simulated and real users on text data confirm that prior elicitation of feature similarities improves prediction accuracy. Furthermore, elicitation with an interactive scatterplot display outperforms straightforward elicitation where the users choose feature pairs from a feature list.

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。

↑