This paper considers the problem of clustering a partially observed unweighted graph---i.e., one where for some node pairs we know there is an edge between them, for some others we know there is no edge, and for the remaining we do not know whether or not there is an edge. We want to organize the nodes into disjoint clusters so that there is relatively dense (observed) connectivity within clusters, and sparse across clusters.
We take a novel yet natural approach to this problem, by focusing on finding the clustering that minimizes the number of "disagreements"---i.e., the sum of the number of (observed) missing edges within clusters, and (observed) present edges across clusters. Our algorithm uses convex optimization; its basis is a reduction of disagreement minimization to the problem of recovering an (unknown) low-rank matrix and an (unknown) sparse matrix from their partially observed sum. We evaluate the performance of our algorithm on the classical Planted Partition/Stochastic Block Model. Our main theorem provides sufficient conditions for the success of our algorithm as a function of the minimum cluster size, edge density and observation probability; in particular, the results characterize the tradeoff between the observation probability and the edge density gap. When there are a constant number of clusters of equal size, our results are optimal up to logarithmic factors.
New York University School of Medicine(纽约大学医学院);Merck Research Laboratories(默克研究实验室);University of Miami School of Business Administration(迈阿密大学工商管理学院);New Jersey Institute of Technology(新泽西理工学院)
Distribution free optimality intervals for clustering
University of Washington(华盛顿大学)
我们解决的问题,验证聚类算法的输出。给定数据$\mathcal{D}$和将这些数据划分为$K$簇的分区$\mathcal{C}$,我们什么时候可以说获得的簇对数据是正确的或有意义的?本文介绍了一种范式,其中聚类$\mathcal{C}$被认为是有意义的,如果它是好的损失函数,如K-均值失真,稳定,即只有良好的聚类小扰动。此外,我们提出了一个通用的方法来获得后推理保证的聚类$\mathcal{C}$的近最优性和稳定性。该方法可以实例化的各种聚类标准(也称为损失函数),凸松弛存在。获得保证相当于解决凸优化问题。我们证明了这种方法的实际相关性,通过获得保证的K-均值和标准化切割聚类标准的现实数据集。我们还证明了渐近不稳定性蕴含有限样本不稳定性w. h. p.,允许从样本推断群体聚类性。保证不依赖于任何分布的假设,但它们依赖于数据集$\mathcal{D}$承认一个稳定的聚类。
A New Framework for Convex Clustering in Kernel Spaces: Finite Sample Bounds, Consistency and Performance Insights
Indian Statistical Institute, Kolkata(印度统计研究院,加尔各答);Electronics and Communication Sciences Unit, Indian Statistical Institute(印度统计研究院电子与通信科学单位);Department of Statistics, University of Oxford(牛津大学统计系);Department of Statistics, University of Michigan(密歇根大学统计系)
Convex Clustering: Model, Theoretical Guarantee and Efficient Algorithm
The Hong Kong Polytechnic University(香港理工大学);National University of Singapore(新加坡国立大学)
聚类是无监督学习中的一个基本问题。像K-means这样的流行方法可能会表现不佳,因为它们容易陷入局部最小值。最近,在Pelckmans等人(2005)、Lindsten等人(2011)和Hocking等人(2011)中提出了范数和(SON)模型(也称为聚类路径)。具有均匀加权所有成对差异正则化的凸聚类模型的完美恢复特性已由Zhu et al.(2014)和Panahi et al.(2017)证明。然而,一般的加权凸聚类模型,其中更好的经验结果已被观察到,没有理论上的保证。在数值优化方面,虽然已经提出了交替方向乘法(ADMM)和交替最小化算法(AMA)等算法来解决凸聚类模型(Chi和Lange,2015),但解决大规模问题仍然非常具有挑战性。本文建立了一般加权凸聚类模型的完全恢复保证的充分条件,作为特例包含和改进了已有的理论结果。此外,我们开发了一个半光滑牛顿的增广拉格朗日方法来解决大规模的凸聚类问题。模拟和真实数据的大量数值实验表明,我们的算法是高效和强大的解决大规模问题。此外,数值结果也显示了优越的性能和可扩展性,我们的算法相比,现有的一阶方法。特别地,我们的算法能够在大约6分钟内解决$\mathbb{R}^3$中具有200,000个点的凸聚类问题。
Clustering by Orthogonal NMF Model and Non-Convex Penalty Optimization
Shenzhen Research Institute of Big Data(深圳大数据研究院);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));University of Minnesota(明尼苏达大学);University of Southern California(南加州大学)
A Convex Formulation for Spectral Shrunk Clustering
University of Technology Sydney(悉尼科技大学);Northwestern Polytechnical University(西北工业大学);University of Texas at Arlington(德克萨斯大学阿灵顿分校);Carnegie Mellon University(卡内基梅隆大学);The University of Queensland(昆士兰大学)
Convex Clustering Redefined: Robust Learning with the Median of Means Estimator
Indian Statistical Institute, Kolkata(印度统计研究院,科希马);Department of Statistics and Data Science, University of Texas at Austin(德克萨斯大学奥斯汀分校统计与数据科学系);Electronics and Communication Sciences, Indian Statistical Institute, Kolkata(印度统计研究院,科希马电子与通信科学);Department of Statistics, University of Oxford(牛津大学统计系);Department of Statistics, University of Michigan(密歇根大学统计系)
Myongji University(明知大学);University of Texas Medical Branch(德克萨斯大学医学分部);Sejong University(世宗大学);OCST Co., Ltd.(OCST有限公司);HCMC University of Technology and Education(胡志明市工业大学)
Multi-Prototypes Convex Merging Based K-Means Clustering Algorithm
Xidian University(西安电子科技大学);The Chinese University of Hong Kong(香港中文大学);City University of Hong Kong(香港城市大学);Hong Kong Centre for Cerebro-Cardiovascular Health Engineering(香港心脑血管健康工程中心)
A low-rank non-convex norm method for multiview graph clustering
The UM6P Vanguard Center, Mohammed VI Polytechnic University, Green City, Morocco(摩洛哥绿色城穆莱·易斯二世理工大学UM6P先锋中心);Université du Littoral Cote d’Opale, LMPA(滨海阿尔卑斯大学LMPA)
When Graph Structure Provably Helps Classification: Non-Asymptotic Recovery Guarantees
Department of Computer Science and Engineering(计算机科学与工程系);Chalmers University of Technology(楚德大学);University of Gothenburg(哥德堡大学);Department of Electrical Engineering(电气工程系)
SMART: Semantic Matching Contrastive Learning for Partially View-Aligned Clustering
College of Computer Science and Technology, Huaqiao University(华侨大学计算机科学与技术学院);Department of Computer Science, Shantou University(汕头大学计算机科学系);Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing, College of Electronic and Information Engineering, Southwest University(西南大学电子信息工程学院非线性电路与智能信息处理重点实验室);School of Computer Science and Engineering, South China University of Technology(华南理工大学计算机科学与工程学院);Department of Computer Science, City University of Hong Kong(香港城市大学计算机科学系)
Partially Observable Markov Decision Process Modelling for Assessing Hierarchies
Insight Centre for Data Analytics(洞察数据分析中心);School of Computer Science, University College Dublin(都柏林大学学院计算机科学学院);Data Science Institute, National University of Ireland, Galway(爱尔兰国立大学戈尔韦数据科学研究所)