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 2405.14532stat.MLcs.LGmath.STstat.TH
收藏

对齐嵌入与几何随机图:Procrustes-Wasserstein问题的信息论结果与计算方法

Aligning Embeddings and Geometric Random Graphs: Informational Results and Computational Approaches for the Procrustes-Wasserstein Problem

DI ENS(巴黎高等师范学院计算科学系);PSL University(巴黎文理研究大学);Inria Paris(法国国家信息与自动化研究所巴黎分部);Université Paris-Saclay(巴黎萨克雷大学);LMO(奥赛数学实验室);MSR-Inria Joint Centre(微软研究院-法国国家信息与自动化研究所联合中心)

Mathieu Even, Luca Ganassali, Jakob Maier, Laurent Massoulié

AI总结:

本文研究Procrustes-Wasserstein问题,提出Ping-Pong算法,通过交替估计正交变换和重标记来对齐高维点云,并给出信息论界限和实验验证。

摘要:

Procrustes-Wasserstein问题旨在无监督设置下匹配两个高维点云,并在自然语言处理和计算机视觉中有许多应用。我们考虑一个植入模型,其中两个数据集$X$和$Y$由$\nmathbb{R}^d$中的$n$个数据点组成,$Y$是$X$的噪声版本,经过正交变换和数据点重新标记。该设置与几何模型中的图对齐问题相关。在本工作中,我们关注点云之间的欧几里得传输代价作为对齐性能的度量。我们首先建立信息论结果,分别在高维($d \gg \log n$)和低维($d \ll \log n$)区域。然后我们研究计算方面,并提出Ping-Pong算法,交替估计正交变换和重新标记,通过Franke-Wolfe凸松弛初始化。我们给出了该方法在单步后检索植入信号的充分条件。我们提供了实验结果,将所提出的方法与Grave等人(2019)的最先进方法进行比较。

英文摘要:

The Procrustes-Wasserstein problem consists in matching two high-dimensional point clouds in an unsupervised setting, and has many applications in natural language processing and computer vision. We consider a planted model with two datasets $X,Y$ that consist of $n$ datapoints in $\mathbb{R}^d$, where $Y$ is a noisy version of $X$, up to an orthogonal transformation and a relabeling of the data points. This setting is related to the graph alignment problem in geometric models. In this work, we focus on the euclidean transport cost between the point clouds as a measure of performance for the alignment. We first establish information-theoretic results, in the high ($d \gg \log n$) and low ($d \ll \log n$) dimensional regimes. We then study computational aspects and propose the Ping-Pong algorithm, alternatively estimating the orthogonal transformation and the relabeling, initialized via a Franke-Wolfe convex relaxation. We give sufficient conditions for the method to retrieve the planted signal after one single step. We provide experimental results to compare the proposed approach with the state-of-the-art method of Grave et al. (2019).

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。

↑