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 2008.06716cs.IRcs.LGstat.ML
收藏

双曲几何模型在Top-N推荐任务上的性能

Performance of Hyperbolic Geometry Models on Top-N Recommendation Tasks

Skolkovo Institute of Science and Technology(斯科尔科沃科学技术研究院);Institute of Numerical Mathematics, Russian Academy of Sciences(俄罗斯科学院数值数学研究所);New York University(纽约大学)

Leyla Mirvakhabova, Evgeny Frolov, Valentin Khrulkov, Ivan Oseledets, Alexander Tuzhilin

摘要:

我们介绍了一个简单的自动编码器的基础上双曲几何解决标准的协同过滤问题。与许多现代深度学习技术相比,我们只使用单个隐藏层来构建解决方案。值得注意的是,即使有这样一个最低限度的方法,我们不仅优于欧几里德对应,但也实现了竞争力的性能相对于目前的最先进的。我们还探讨了空间曲率的影响双曲模型的质量,并提出了一个有效的数据驱动的方法来估计其最佳值。

英文摘要:

We introduce a simple autoencoder based on hyperbolic geometry for solving standard collaborative filtering problem. In contrast to many modern deep learning techniques, we build our solution using only a single hidden layer. Remarkably, even with such a minimalistic approach, we not only outperform the Euclidean counterpart but also achieve a competitive performance with respect to the current state-of-the-art. We additionally explore the effects of space curvature on the quality of hyperbolic models and propose an efficient data-driven method for estimating its optimal value.

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。

↑