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 2005.03932cs.IRcs.LG
收藏

基于正则化自注意的文档交互学习排序模型

Modeling Document Interactions for Learning to Rank with Regularized Self-Attention

Johns Hopkins University(约翰斯·霍普金斯大学)

Shuo Sun, Kevin Duh

摘要:

学习排序是一个重要的任务,已经成功地部署在许多现实世界的信息检索系统。大多数现有的方法计算独立的文档的相关性判断,没有整体考虑整个竞争文件集。在本文中,我们将探讨基于自注意的神经网络的文档交互建模。虽然自我注意网络在许多NLP任务中取得了最先进的成果,但我们发现,经验表明,自我注意与基线神经学习排序架构相比,几乎没有什么好处。为了改进自注意权重的学习,我们提出了简单而有效的正则化项,旨在对文档之间的交互进行建模。在公开的学习排序(LETOR)数据集上的评估表明,使用我们提出的正则化项训练自注意网络可以显着优于现有的学习排序方法。

英文摘要:

Learning to rank is an important task that has been successfully deployed in many real-world information retrieval systems. Most existing methods compute relevance judgments of documents independently, without holistically considering the entire set of competing documents. In this paper, we explore modeling documents interactions with self-attention based neural networks. Although self-attention networks have achieved state-of-the-art results in many NLP tasks, we find empirically that self-attention provides little benefit over baseline neural learning to rank architecture. To improve the learning of self-attention weights, We propose simple yet effective regularization terms designed to model interactions between documents. Evaluations on publicly available Learning to Rank (LETOR) datasets show that training self-attention network with our proposed regularization terms can significantly outperform existing learning to rank methods.

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。

↑