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 1711.09446cs.IR
收藏

在线学习中信息检索排序速度与质量的平衡

Balancing Speed and Quality in Online Learning to Rank for Information Retrieval

Harrie Oosterhuis, Maarten de Rijke

摘要:

在在线学习排序(OLTR)的目的是找到一个最佳的排名模型,通过与用户的交互。当从用户行为中学习时,系统必须与用户交互,同时从这些交互中学习。与其他学习排序(LTR)设置不同,该领域的现有研究仅限于线性模型。这是由于选择模型时出现的速度-质量权衡:复杂的模型更具表达力,可以找到最佳排名,但需要更多的用户交互才能做到这一点,这一要求可能会在训练过程中让用户感到沮丧。相反,更简单的模型可以在更少的交互上进行优化,从而提供更好的用户体验,但它们将收敛于次优排名。这种权衡造成了僵局,因为新的模型将无法在不牺牲另一个的情况下改善用户体验或最终的收敛点。我们的贡献是双重的。首先,我们介绍了一个快速OLTR模型称为Sim-MGD,解决了速度方面的速度质量权衡。Sim-MGD基于与参考文档的相似性对文档进行排名。它收敛迅速,因此,提供了一个更好的用户体验,但它不收敛到最佳排名。其次,我们贡献了级联多叶梯度下降(C-MGD)的OLTR,直接解决了速度和质量的权衡使用级联,使最好的两个世界的组合:快速学习和高质量的最终收敛。C-MGD可以提供Sim-MGD的更好的用户体验,同时保持与最先进的MGD模型相同的收敛。这为未来的工作打开了大门,为OLTR设计新的模型,而不必处理速度-质量权衡。

英文摘要:

In Online Learning to Rank (OLTR) the aim is to find an optimal ranking model by interacting with users. When learning from user behavior, systems must interact with users while simultaneously learning from those interactions. Unlike other Learning to Rank (LTR) settings, existing research in this field has been limited to linear models. This is due to the speed-quality tradeoff that arises when selecting models: complex models are more expressive and can find the best rankings but need more user interactions to do so, a requirement that risks frustrating users during training. Conversely, simpler models can be optimized on fewer interactions and thus provide a better user experience, but they will converge towards suboptimal rankings. This tradeoff creates a deadlock, since novel models will not be able to improve either the user experience or the final convergence point, without sacrificing the other. Our contribution is twofold. First, we introduce a fast OLTR model called Sim-MGD that addresses the speed aspect of the speed-quality tradeoff. Sim-MGD ranks documents based on similarities with reference documents. It converges rapidly and, hence, gives a better user experience but it does not converge towards the optimal rankings. Second, we contribute Cascading Multileave Gradient Descent (C-MGD) for OLTR that directly addresses the speed-quality tradeoff by using a cascade that enables combinations of the best of two worlds: fast learning and high quality final convergence. C-MGD can provide the better user experience of Sim-MGD while maintaining the same convergence as the state-of-the-art MGD model. This opens the door for future work to design new models for OLTR without having to deal with the speed-quality tradeoff.

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。

↑