arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2204.04670cs.LG
收藏

带标签比较的主动学习

Active Learning with Label Comparisons

Google(谷歌);Weizmann Institute of Science(魏茨曼科学研究所);Technion(以色列理工学院);Hebrew University(希伯来大学);Tel Aviv University(特拉维夫大学)

Gal Yona, Shay Moran, Gal Elidan, Amir Globerson

摘要:

监督学习通常依赖于真实标签的手动注释。当有许多潜在的类时,搜索最好的类对于注释人员来说是令人望而却步的。另一方面,比较两个候选标签通常要容易得多。我们专注于这种类型的成对监督,并询问它如何有效地用于学习,特别是在主动学习。在此背景下,我们得到了一些有见地的结果。原则上,可以使用$k-1$主动查询来找到最佳的$k$标签。我们发现,有一个自然的类,这种方法是次优的,并且有一个比较有效的主动学习计划。我们分析中的一个关键要素是真实分布的“标签邻域图”,如果两个类共享一个决策边界,则它们之间有一个边缘。我们还表明,在PAC设置,成对比较不能提供改进的样本复杂性在最坏的情况下。我们用实验来补充我们的理论结果,清楚地表明了邻域图对样本复杂性的影响。

英文摘要:

Supervised learning typically relies on manual annotation of the true labels. When there are many potential classes, searching for the best one can be prohibitive for a human annotator. On the other hand, comparing two candidate labels is often much easier. We focus on this type of pairwise supervision and ask how it can be used effectively in learning, and in particular in active learning. We obtain several insightful results in this context. In principle, finding the best of $k$ labels can be done with $k-1$ active queries. We show that there is a natural class where this approach is sub-optimal, and that there is a more comparison-efficient active learning scheme. A key element in our analysis is the "label neighborhood graph" of the true distribution, which has an edge between two classes if they share a decision boundary. We also show that in the PAC setting, pairwise comparisons cannot provide improved sample complexity in the worst case. We complement our theoretical results with experiments, clearly demonstrating the effect of the neighborhood graph on sample complexity.

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。

↑