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 2103.05290stat.MLcs.LG
收藏

基于先验的近似隐黎曼度量

A prior-based approximate latent Riemannian metric

Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);Fraunhofer IAIS(弗劳恩霍夫智能分析与信息系统研究所)

Georgios Arvanitidis, Bogdan Georgiev, Bernhard Schölkopf

摘要:

随机生成模型使我们能够通过潜在空间中的黎曼度量来捕获位于高维空间中的数据流形的几何结构。然而,它的实际应用是相当有限的,主要是由于不可避免的复杂性。在这项工作中,我们提出了一个替代共形黎曼度量的生成模型,是简单,有效和强大的潜在空间。这个指标是基于一个可学习的先验知识,我们建议使用一个基本的基于能量的模型来学习。我们从理论上分析了所提出的度量的行为,并表明它是明智的,在实践中使用。我们证明实验的效率和鲁棒性,以及新的近似度量的行为。此外,我们显示了所提出的方法在生命科学中的数据分析的适用性。

英文摘要:

Stochastic generative models enable us to capture the geometric structure of a data manifold lying in a high dimensional space through a Riemannian metric in the latent space. However, its practical use is rather limited mainly due to inevitable complexity. In this work we propose a surrogate conformal Riemannian metric in the latent space of a generative model that is simple, efficient and robust. This metric is based on a learnable prior that we propose to learn using a basic energy-based model. We theoretically analyze the behavior of the proposed metric and show that it is sensible to use in practice. We demonstrate experimentally the efficiency and robustness, as well as the behavior of the new approximate metric. Also, we show the applicability of the proposed methodology for data analysis in the life sciences.

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。

↑