Experimental Design under the Bradley-Terry Model

Experimental Design under the Bradley-Terry Model
复制标题

Bradley-Terry 模型下的实验设计

DOI:
10.24963/ijcai.2018/304
复制
发表时间:
2018
期刊:
Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems
影响因子:
--
通讯作者:
Stratis Ioannidis
Stratis Ioannidis
中科院分区:
--
文献类型:
--
作者:
Yuan Guo;Peng Tian;Jayashree Kalpathy;S. Ostmo;J. Campbell;M. Chiang;Deniz Erdoğmuş;Jennifer G. Dy;Stratis Ioannidis

文献摘要

参考文献

被引文献

相似文献

与传统样本标签相比,人类专家通过比较生成的标签显示出较小的差异。收集比较标签在大型数据集上具有挑战性,因为比较的数量随数据集大小的次数增长。我们研究以下实验设计问题:鉴于专家比较的预算,以及一组现有样本标签,我们确定了比较标签,以收集导致最高分类的改进。我们研究了Bradley-Terry模型动机的几个实验设计目标。由此产生的优化问题等于最大程度地提高了下函数功能。我们通过实验评估这些方法在合成和现实生活数据集上的性能。
Labels generated by human experts via comparisons exhibit smaller variance compared to traditional sample labels. Collecting comparison labels is challenging over large datasets, as the number of comparisons grows quadratically with the dataset size. We study the following experimental design problem: given a budget of expert comparisons, and a set of existing sample labels, we determine the comparison labels to collect that lead to the highest classification improvement. We study several experimental design objectives motivated by the Bradley-Terry model. The resulting optimization problems amount to maximizing submodular functions. We experimentally evaluate the performance of these methods over synthetic and real-life datasets.
DOI: 10.1056/nejmra1208129
发表时间: 2012-12-27
期刊: The New England journal of medicine
影响因子: --
作者:
Hartnett ME;Penn JS
通讯作者: Penn JS
DOI: 10.1016/j.jspi.2010.04.025
发表时间: 2010-12-01
影响因子: 0.9
作者:
Wahba, Grace
通讯作者: Wahba, Grace