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 1909.02768cs.IRcs.LGstat.ML
收藏

基于神经网络的两两学习排序:重构、理论分析与实际性能

Pairwise Learning to Rank by Neural Networks Revisited: Reconstruction, Theoretical Analysis and Practical Performance

Johannes Gutenberg-Universität Mainz(美因茨约翰内斯·古腾堡大学);University of Birmingham(伯明翰大学)

Marius Köppel, Alexander Segner, Martin Wagener, Lukas Pensel, Andreas Karwath, Stefan Kramer

摘要:

我们提出了一个成对的学习排序方法的基础上的神经网络,称为DirectRanker,概括的RankNet架构。我们在数学上证明了我们的模型是自反的,反对称的,并且是传递的,允许简化训练和提高性能。在LETOR MSLR-WEB 10 K、MQ 2007和MQ 2008数据集上的实验结果表明,我们的模型优于许多最先进的方法,同时在结构上更简单,并且只使用成对方法。

英文摘要:

We present a pairwise learning to rank approach based on a neural net, called DirectRanker, that generalizes the RankNet architecture. We show mathematically that our model is reflexive, antisymmetric, and transitive allowing for simplified training and improved performance. Experimental results on the LETOR MSLR-WEB10K, MQ2007 and MQ2008 datasets show that our model outperforms numerous state-of-the-art methods, while being inherently simpler in structure and using a pairwise approach only.

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。

↑