arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.07400stat.MLcs.IRcs.LG
收藏

直接学习排名和重新排名

Direct Learning to Rank and Rerank

Duke University(杜克大学);Carnegie Mellon University(卡内基梅隆大学)

Cynthia Rudin, Yining Wang

摘要:

学习排名技术已被证明是非常有用的优先级问题,我们排名项目的顺序,其估计的概率,并致力于我们有限的资源排名靠前的项目。这项工作暴露了一个严重的问题,学习排名算法的状态,这是他们是基于凸代理,导致穷人的近似。然后,我们讨论了“精确”重新排序算法的数学规划的基础上的可能性。我们证明了一个放松版本的“精确”问题有相同的最优解,并提供了一个实证分析。

英文摘要:

Learning-to-rank techniques have proven to be extremely useful for prioritization problems, where we rank items in order of their estimated probabilities, and dedicate our limited resources to the top-ranked items. This work exposes a serious problem with the state of learning-to-rank algorithms, which is that they are based on convex proxies that lead to poor approximations. We then discuss the possibility of "exact" reranking algorithms based on mathematical programming. We prove that a relaxed version of the "exact" problem has the same optimal solution, and provide an empirical analysis.

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。

↑