arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.03060cs.IRcs.LG
收藏

基于多目标优化的多标记学习排序

Multi-Label Learning to Rank through Multi-Objective Optimization

National University of Singapore(新加坡国立大学);Amazon.com Inc(亚马逊公司)

Debabrata Mahapatra, Chaosheng Dong, Yetian Chen, Deqiang Meng, Michinari Momma

摘要:

学习排序(Learning to Rank,LTR)技术在当今的信息检索系统中普遍存在,尤其是在搜索排序应用中。通常用于训练排名模型的查询项相关性标签通常是人类行为的噪声测量,例如,产品搜索的产品评级。粗略测量使得地面实况排名相对于单个相关性准则是非唯一的。为了解决模糊性,需要使用许多相关性准则来训练模型,从而产生多标签LTR(MLLTR)。此外,它制定了多个目标,这些目标可能是冲突的,但同时优化是重要的,例如,在产品搜索中,可以基于产品质量和购买可能性来训练排名模型以增加收入。在这项研究中,我们利用多目标优化(MOO)方面的MLLTR问题,并采用最近开发的MOO算法来解决它。具体来说,我们提出了一个通用的框架,标签的信息可以结合在各种方式来有意义地表征目标之间的权衡。我们的框架允许任何基于梯度的MOO算法用于解决MLLTR问题。我们在两个公开的LTR数据集和一个电子商务数据集上测试了所提出的框架,以显示其有效性。

英文摘要:

Learning to Rank (LTR) technique is ubiquitous in the Information Retrieval system nowadays, especially in the Search Ranking application. The query-item relevance labels typically used to train the ranking model are often noisy measurements of human behavior, e.g., product rating for product search. The coarse measurements make the ground truth ranking non-unique with respect to a single relevance criterion. To resolve ambiguity, it is desirable to train a model using many relevance criteria, giving rise to Multi-Label LTR (MLLTR). Moreover, it formulates multiple goals that may be conflicting yet important to optimize for simultaneously, e.g., in product search, a ranking model can be trained based on product quality and purchase likelihood to increase revenue. In this research, we leverage the Multi-Objective Optimization (MOO) aspect of the MLLTR problem and employ recently developed MOO algorithms to solve it. Specifically, we propose a general framework where the information from labels can be combined in a variety of ways to meaningfully characterize the trade-off among the goals. Our framework allows for any gradient based MOO algorithm to be used for solving the MLLTR problem. We test the proposed framework on two publicly available LTR datasets and one e-commerce dataset to show its efficacy.

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。

↑