Filters








135 Hits in 2.9 sec

Probabilistic Clustering of Time-Evolving Distance Data [article]

Julia E. Vogt, Marius Kloft, Stefan Stark, Sudhir S. Raman, Sandhya Prabhakaran, Volker Roth, Gunnar Rätsch
2015 arXiv   pre-print
Further, the model does not require the number of clusters being specified in advance -- they are instead determined automatically using a Dirichlet process prior.  ...  We present a novel probabilistic clustering model for objects that are represented via pairwise distances and observed at different time points.  ...  A Time-evolving Translation-invariant Wishart-Dirichlet Process In this section, we present a novel dynamic clustering approach, the time-evolving translation-invariant Wishart-Dirichlet process (Te-TiWD  ... 
arXiv:1504.03701v1 fatcat:cp54cg3em5erdk4asv7ldorrtm

Probabilistic clustering of time-evolving distance data

Julia E. Vogt, Marius Kloft, Stefan Stark, Sudhir S. Raman, Sandhya Prabhakaran, Volker Roth, Gunnar Rätsch
2015 Machine Learning  
Further, the model does not require the number of clusters being specified in advance-they are instead determined automatically using a Dirichlet process prior.  ...  We present a novel probabilistic clustering model for objects that are represented via pairwise distances and observed at different time points.  ...  Access to patient data is covered under IRB Waiver #WA0426-13.  ... 
doi:10.1007/s10994-015-5516-x fatcat:oiisxrevb5fqfjd2w6qvr677s4

Dirichlet process mixture models on symmetric positive definite matrices for appearance clustering in video surveillance applications

Anoop Cherian, Vassilios Morellas, Nikolaos Papanikolopoulos, Saad J. Bedros
2011 CVPR 2011  
In this paper, a novel application of the Dirichlet Process Mixture Model framework is proposed towards unsupervised clustering of symmetric positive definite matrices.  ...  Since these matrices do not adhere to the Euclidean geometry, clustering algorithms using the traditional distance measures cannot be directly extended to them.  ...  Arindam Banerjee (University of Minnesota) for many helpful discussions. This material is based upon work supported in part by the  ... 
doi:10.1109/cvpr.2011.5995723 dblp:conf/cvpr/CherianMPB11 fatcat:vznh2fzqj5a4hgzwshxherbxgq

Recovering networks from distance data

Sandhya Prabhakaran, David Adametz, Karin J. Metzner, Alexander Böhm, Volker Roth
2013 Machine Learning  
A fully probabilistic approach to reconstructing Gaussian graphical models from distance data is presented.  ...  The main idea is to extend the usual central Wishart model in traditional methods to using a likelihood depending only on pairwise distances, thus being independent of geometric assumptions about the underlying  ...  The likelihood model invariant to shifts has been studied before in the Translational-invariant Wishart Dirichlet (TiWD) cluster process (Vogt et al. 2010 ).  ... 
doi:10.1007/s10994-013-5370-7 fatcat:m6soaxrd7jfadhnixn52jcibmu

Bayesian Clustering of Shapes of Curves [article]

Zhengwu Zhang, Debdeep Pati, Anuj Srivastava
2015 arXiv   pre-print
The elastic-inner product matrix obtained from the data is modeled using a Wishart distribution whose parameters are assigned carefully chosen prior distributions to allow for automatic inference on the  ...  Posterior is sampled through an efficient Markov chain Monte Carlo procedure based on the Chinese restaurant process to infer (1) the posterior distribution on the number of clusters, and (2) clustering  ...  This prior is derived from a Dirichlet process (realized using the Chinese restaurant process) and the likelihood is given by the Wishart distribution.  ... 
arXiv:1504.00377v1 fatcat:475e4sqarbgafo3hqhcrf24zuq

Bayesian clustering of shapes of curves

Zhengwu Zhang, Debdeep Pati, Anuj Srivastava
2015 Journal of Statistical Planning and Inference  
The elastic-inner product matrix obtained from the data is modeled using a Wishart distribution whose parameters are assigned carefully chosen prior distributions to allow for automatic inference on the  ...  Posterior is sampled through an efficient Markov chain Monte Carlo procedure based on the Chinese restaurant process to infer (1) the posterior distribution on the number of clusters, and (2) clustering  ...  This prior is derived from a Dirichlet process (realized using the Chinese restaurant process) and the likelihood is given by the Wishart distribution.  ... 
doi:10.1016/j.jspi.2015.04.007 fatcat:4mgroxjfgrfwrc6tinrhraaknm

Bayesian nonparametric clustering in phylogenetics: modeling antigenic evolution in influenza

Gabriela B. Cybis, Janet S. Sinsheimer, Trevor Bedford, Andrew Rambaut, Philippe Lemey, Marc A. Suchard
2017 Statistics in Medicine  
We propose a phylogenetic Chinese restaurant process that extends the current process to incorporate the phylogenetic dependency structure between strains in the modeling of antigenic clusters.  ...  Analysis of such assay data ideally leads to natural clustering of influenza strains of similar antigenicity that correlate with sequence evolution.  ...  Acknowledgements The research leading to these results has received funding from the European  ... 
doi:10.1002/sim.7196 pmid:28098392 pmcid:PMC5515700 fatcat:vlaaf4ka2rh25aodtwgpzfn24m

Dirichlet process mixtures under affine transformations of the data [article]

Julyan Arbel, Riccardo Corradin, Bernardo Nipoti
2020 arXiv   pre-print
Location-scale Dirichlet process mixtures of Gaussians (DPM-G) have proved extremely useful in dealing with density estimation and clustering problems in a wide range of domains.  ...  First, we devise a coherent prior specification of the model which makes posterior inference invariant with respect to affine transformations of the data.  ...  Acknowledgements This work was developed in the framework of the Ulysses Program for French-Irish collaborations (43135ZK) and the Grenoble Alpes Data Institute.  ... 
arXiv:1809.02463v3 fatcat:u3lmry6shnc2bf3yh5lzuhbthe

Describing Visual Scenes Using Transformed Objects and Parts

Erik B. Sudderth, Antonio Torralba, William T. Freeman, Alan S. Willsky
2007 International Journal of Computer Vision  
The resulting transformed Dirichlet process (TDP) leads to Monte Carlo algorithms which simultaneously segment and recognize objects in street and office scenes.  ...  Turning to multiple object scenes, we propose nonparametric models which use Dirichlet processes to automatically learn the number of parts underlying each object category, and objects composing each scene  ...  Acknowledgements The authors thank Josh Tenenbaum, Daniel Huttenlocher, and the anonymous reviewers for helpful suggestions.  ... 
doi:10.1007/s11263-007-0069-5 fatcat:gcwozuh44rapbmufuewiqxdmhe

Dirichlet process mixtures under affine transformations of the data

Julyan Arbel, Riccardo Corradin, Bernardo Nipoti
2020 Computational statistics (Zeitschrift)  
Location-scale Dirichlet process mixtures of Gaussians (DPM-G) have proved extremely useful in dealing with density estimation and clustering problems in a wide range of domains.  ...  First, we devise a coherent prior specification of the model which makes posterior inference invariant with respect to affine transformations of the data.  ...  This work was developed in the framework of the Ulysses Program for French-Irish collaborations (43135ZK) and the Grenoble Alpes Data Institute.  ... 
doi:10.1007/s00180-020-01013-y fatcat:fvm33vg4rrgxpohjnfnxcr7stm

A sticky HDP-HMM with application to speaker diarization

Emily B. Fox, Erik B. Sudderth, Michael I. Jordan, Alan S. Willsky
2011 Annals of Applied Statistics  
To scale the resulting architecture to realistic diarization problems, we develop a sampling algorithm that employs a truncated approximation of the Dirichlet process to jointly resample the full state  ...  To address this problem, we take a Bayesian nonparametric approach to speaker diarization that builds on the hierarchical Dirichlet process hidden Markov model (HDP-HMM) of Teh et al. [J. Amer.  ...  Morgan for helpful discussions about the NIST dataset.  ... 
doi:10.1214/10-aoas395 fatcat:2ffqxbaocjf5hapn6zjkgexf3e

Capsule Routing via Variational Bayes [article]

Fabio De Sousa Ribeiro, Georgios Leontidis, Stefanos Kollias
2019 arXiv   pre-print
The relationships between objects and their parts are learned via trainable viewpoint-invariant transformation matrices, and the presence of a given object is decided by the level of agreement among votes  ...  This interaction occurs between capsule layers and is a process called routing-by-agreement.  ...  However, capsule routing does differ from regular clustering substantially, as every cluster has its own learnable viewpoint-invariant transformation matrix W ij with which it transforms its data points  ... 
arXiv:1905.11455v3 fatcat:fz7lck4gpvchtjwq25ktuw3uoq

Identifying Mixtures of Mixtures Using Bayesian Estimation

Gertraud Malsiner-Walli, Sylvia Frühwirth-Schnatter, Bettina Grün
2017 Journal of Computational And Graphical Statistics  
The use of a finite mixture of normal distributions in model-based clustering allows to capture non-Gaussian data clusters.  ...  However, identifying the clusters from the normal components is challenging and in general either achieved by imposing constraints on the model or by using post-processing procedures.  ...  Grün gratefully acknowledges support by the Austrian Science Fund (FWF): V170-N18.  ... 
doi:10.1080/10618600.2016.1200472 pmid:28626349 pmcid:PMC5455957 fatcat:tohctqpnyjduti4ium7zjrq2hy

Dual Sticky Hierarchical Dirichlet Process Hidden Markov Model and Its Application to Natural Language Description of Motions

Weiming Hu, Guodong Tian, Yongxin Kang, Chunfeng Yuan, Stephen Maybank
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
In this paper, a new nonparametric Bayesian model called the dual sticky hierarchical Dirichlet process hidden Markov model (HDP-HMM) is proposed for mining activities from a collection of time series  ...  All the time series data are clustered. Each cluster of time series data, corresponding to a motion pattern, is modeled by an HMM.  ...  The PHoG is invariant under spatial translations.  ... 
doi:10.1109/tpami.2017.2756039 pmid:28952936 fatcat:6lyk2y3x3zforcjvqzlg7n4fba

Learning hierarchical models of scenes, objects, and parts

E.B. Sudderth, A. Torralba, W.T. Freeman, A.S. Willsky
2005 Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1  
We use K-means clustering to  ...  We describe a hierarchical probabilistic model for the detection and recognition of objects in cluttered, natural scenes.  ...  Acknowledgments Funding provided by the National Geospatial-Intelligence Agency NEGI-1582-04-0004, the National Science Foundation NSF-IIS-0413232, the ARDA VACE program, and a BAE Systems grant.  ... 
doi:10.1109/iccv.2005.137 dblp:conf/iccv/SudderthTFW05 fatcat:qh7xirl4qnga5l5ixqkqdhdnam
« Previous Showing results 1 — 15 out of 135 results