arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.08789math.APcs.LGcs.NAmath.NA
收藏

图上的Hamilton-Jacobi方程及其在半监督学习和数据深度中的应用

Hamilton-Jacobi equations on graphs with applications to semi-supervised learning and data depth

University of Minnesota(明尼苏达大学)

Jeff Calder, Mahmood Ettehad

摘要:

最短路径图距离广泛用于数据科学和机器学习,因为它们可以近似数据流形上的底层测地线距离。然而,最短路径距离对图中添加损坏的边非常敏感,无论是通过噪声还是对抗性扰动。本文研究了图上的一类Hamilton-Jacobi方程,我们称之为$p$ -程函方程。我们证明了$p$ -程函方程与$p=1$是一个可证明的强大的距离型函数的图,和$p\to \infty$限制恢复最短路径距离。虽然$p$ -程函方程不对应于最短路径图的距离,我们仍然表明,连续极限的$p$ -程函方程的随机几何图恢复测地密度加权距离在连续。我们考虑应用$p$ -程函方程的数据深度和半监督学习,并使用连续极限证明这两个应用程序的渐近一致性结果。最后,我们展示了在真实图像数据集上进行数据深度和半监督学习的实验结果,包括MNIST,FashionMNIST和CIFAR-10,结果表明,与最短路径距离相比,$p$ -程函方程提供了更好的结果。

英文摘要:

Shortest path graph distances are widely used in data science and machine learning, since they can approximate the underlying geodesic distance on the data manifold. However, the shortest path distance is highly sensitive to the addition of corrupted edges in the graph, either through noise or an adversarial perturbation. In this paper we study a family of Hamilton-Jacobi equations on graphs that we call the $p$-eikonal equation. We show that the $p$-eikonal equation with $p=1$ is a provably robust distance-type function on a graph, and the $p\to \infty$ limit recovers shortest path distances. While the $p$-eikonal equation does not correspond to a shortest-path graph distance, we nonetheless show that the continuum limit of the $p$-eikonal equation on a random geometric graph recovers a geodesic density weighted distance in the continuum. We consider applications of the $p$-eikonal equation to data depth and semi-supervised learning, and use the continuum limit to prove asymptotic consistency results for both applications. Finally, we show the results of experiments with data depth and semi-supervised learning on real image datasets, including MNIST, FashionMNIST and CIFAR-10, which show that the $p$-eikonal equation offers significantly better results compared to shortest path distances.

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。

↑