Exemplar-based clustering methods have been shown to produce state-of-the-art results on a number of synthetic and real-world clustering problems. They are appealing because they offer computational benefits over latent-mean models and can handle arbitrary pairwise similarity measures between data points. However, when trying to recover underlying structure in clustering problems, tailored similarity measures are often not enough; we also desire control over the distribution of cluster sizes. Priors such as Dirichlet process priors allow the number of clusters to be unspecified while expressing priors over data partitions. To our knowledge, they have not been applied to exemplar-based models. We show how to incorporate priors, including Dirichlet process priors, into the recently introduced affinity propagation algorithm. We develop an efficient maxproduct belief propagation algorithm for our new model and demonstrate experimentally how the expanded range of clustering priors allows us to better recover true clusterings in situations where we have some information about the generating process.
Explainable Clustering via Exemplars: Complexity and Efficient Approximation Algorithms
Biocomplexity Institute, University of Virginia（弗吉尼亚大学生物复杂性研究所）；University at Albany – State University of New York（纽约州立大学奥尔巴尼分校）；University of California Davis（加州大学戴维斯分校）；University of Lyon（里昂大学）；Joint AI Center (JAIC)（联合人工智能中心（JAIC））
A flexible EM-like clustering algorithm for noisy data
Laboratoire des Signaux et Systèmes (L2S)（信号与系统实验室）；CentraleSupélec-CNRS-Université Paris-Sud（巴黎中央理工-高等电力学院-法国国家科学研究中心-巴黎第十一大学）；Université Paris-Saclay（巴黎萨克雷大学）；Instituto de Cálculo（计算研究所）；Universidad de Buenos Aires（布宜诺斯艾利斯大学）；IMAS-CONICET（阿根廷国家科学技术研究委员会数学应用研究所）
虽然EM for GMM算法非常流行，但众所周知，EM for GMM算法具有非高斯分布形状，离群值和高维性。在本文中，我们设计了一个新的强大的聚类算法，可以有效地处理不同的数据集的噪声和离群点。作为一种类EM算法，它同时基于聚类中心和协方差的估计。此外，使用半参数范例，该方法估计每个数据点的未知尺度参数。这使得算法能够适应较重的尾部分布和离群值，而不会在各种经典场景中显着降低效率。我们首先推导和分析所提出的算法的背景下，椭圆分布，特别是重要的不敏感性的基本数据分布。然后，我们研究的收敛性和精度的算法，考虑第一合成数据。然后，我们表明，该算法优于其他经典的无监督方法的文献，如k-均值，EM高斯混合模型及其最近的修改或谱聚类时，应用到真实的数据集MNIST，NORB，和20新闻组。
Robust Unsupervised Flexible Auto-weighted Local-Coordinate Concept Factorization for Image Clustering
Soochow University（苏州大学）；Hefei University of Technology（合肥工业大学）；Adobe Research（奥多比研究院）；Nanjing University of Information Science and Technology（南京信息工程大学）；National University of Singapore（新加坡国立大学）
A Flexible Adaptive Stable Clustering Algorithm for Archive-Scale Online Mass Spectrometry
Shenzhen Key Laboratory of Precision Measurement and Early Warning Technology for Urban Environmental Health Risks, School of Environmental Science and Engineering, Southern University of Science and Technology（深圳城市环境健康风险精密测量与预警技术重点实验室，环境科学与工程学院，南方科技大学）
Instituto de Matemática Universidade Federal do Rio de Janeiro（里约热内卢联邦大学数学研究所）；Departamento de Métodos Estatísticos Universidade Federal do Rio de Janeiro（里约热内卢联邦大学统计方法系）
