arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2301.03338cs.LG
收藏

拓扑正则化的数据嵌入

Topologically Regularized Data Embeddings

Ghent University(根特大学);VIB Inflammation Research Center(VIB炎症研究中心)

Edith Heiter, Robin Vandaele, Tijl De Bie, Yvan Saeys, Jefrey Lijffijt

AI总结:

针对现有无监督表示学习方法无法在低维嵌入中保留已知拓扑先验的问题,提出基于代数拓扑的拓扑正则化通用方法,通过联合优化嵌入损失与拓扑损失得到兼具局部邻近性与目标拓扑结构的嵌入,并验证了其效率、鲁棒性与通用性。

摘要:

无监督表示学习方法被广泛用于洞察高维、非结构化或结构化数据。在某些场景下,用户可能具备关于数据的先验拓扑知识,例如已知的聚类结构,或是数据沿树状或图结构拓扑分布的事实。然而,目前尚缺乏能确保这类结构在低维表示中显著呈现的通用方法。这会降低低维嵌入的可解释性,还可能对下游学习任务产生不利影响。为解决这一问题,我们提出拓扑正则化方法:一种基于代数拓扑的通用框架,用于将拓扑先验知识融入低维嵌入。我们提出了一类拓扑损失函数,并证明将嵌入损失与这类拓扑损失函数作为正则项联合优化,得到的嵌入不仅能反映局部邻近性,还能呈现期望的拓扑结构。本文包含代数拓扑所需基础概念的自包含概述,并针对如何为聚类、环、分岔等多种形状设计拓扑损失函数提供了直观指导。我们结合线性与非线性降维方法以及图嵌入方法,从计算效率、鲁棒性和通用性三方面对所提方法进行了实证评估。

英文摘要:

Unsupervised representation learning methods are widely used for gaining insight into high-dimensional, unstructured, or structured data. In some cases, users may have prior topological knowledge about the data, such as a known cluster structure or the fact that the data is known to lie along a tree- or graph-structured topology. However, generic methods to ensure such structure is salient in the low-dimensional representations are lacking. This negatively impacts the interpretability of low-dimensional embeddings, and plausibly downstream learning tasks. To address this issue, we introduce topological regularization: a generic approach based on algebraic topology to incorporate topological prior knowledge into low-dimensional embeddings. We introduce a class of topological loss functions, and show that jointly optimizing an embedding loss with such a topological loss function as a regularizer yields embeddings that reflect not only local proximities but also the desired topological structure. We include a self-contained overview of the required foundational concepts in algebraic topology, and provide intuitive guidance on how to design topological loss functions for a variety of shapes, such as clusters, cycles, and bifurcations. We empirically evaluate the proposed approach on computational efficiency, robustness, and versatility in combination with linear and non-linear dimensionality reduction and graph embedding 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。

↑