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 2510.25354cs.LGmath.STstat.TH
收藏

超图上的半监督学习分析

Analysis of Semi-Supervised Learning on Hypergraphs

University of California Los Angeles(加州大学洛杉矶分校);University of Warwick(沃里克大学)

Adrien Weihs, Andrea L. Bertozzi, Matthew Thorpe

AI总结:

研究超图上的半监督学习问题,提出高阶超图学习(HOHL)方法,通过分析随机几何超图上的变分问题,确定缩放机制,证明离散极小值收敛,还给出多尺度拉普拉斯替代并证明收敛,数值实验验证了高阶正则化的实用性。

摘要:

超图为建模多路交互提供了自然框架。我们分析了一类在随机几何超图上提出的变分半监督学习问题,并在大数据极限下建立了渐近一致性。特别地,我们确定了确保适定性的缩放机制,展示了离散极小值在连续统中收敛到密度加权p -拉普拉斯方程的解。还提出了高阶超图学习(HOHL),并分析了其在几何点云上的高效多尺度拉普拉斯替代,数值实验支持了所得高阶正则化的实用性。

英文摘要:

Hypergraphs provide a natural framework for modeling multiway interactions. We analyze a class of variational semi-supervised learning problems posed on random geometric hypergraphs and establish asymptotic consistency in the large-data limit. In particular, we identify scaling regimes that ensure well-posedness--yielding nontrivial label propagation rather than collapse to a constant labeling--and show that discrete minimizers converge, in the continuum, to solutions of a density-weighted p-Laplacian equation. We also propose Higher-Order Hypergraph Learning (HOHL), a multiscale regularization scheme based on powers of Laplacians associated with hypergraph-induced subgraphs. For geometric point clouds, we analyze an efficient multiscale Laplacian surrogate for HOHL and prove convergence to a higher-order Sobolev-type seminorm. Numerical experiments on standard benchmarks support the practical utility of the resulting higher-order regularization.

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。

↑