Filters








8,444 Hits in 8.1 sec

Estimating Multiple Precision Matrices with Cluster Fusion Regularization [article]

Bradley S. Price and Aaron J. Molstad and Ben Sherwood
2020 arXiv   pre-print
We propose a penalized likelihood framework for estimating multiple precision matrices from different classes.  ...  The framework proposed in this article allows for simultaneous estimation of the precision matrices and relationships between the precision matrices, jointly.  ...  In our proposed estimators, the cluster fusion penalty is used to promote similarity in precision matrices that are in the same cluster, while estimating precision matrices in different clusters separately  ... 
arXiv:2003.00371v1 fatcat:6mktdfgupne2vpmc3hm537zae4

Estimating Multiple Precision Matrices with Cluster Fusion Regularization

Bradley S. Price, Aaron J. Molstad, Ben Sherwood
2021 figshare.com  
We propose a penalized likelihood framework for estimating multiple precision matrices from different classes.  ...  The framework proposed in this article allows for simultaneous estimation of the precision matrices and relationships between the precision matrices.  ...  10 3 for the optimal Q in each simulation. precision matrices are dense.  ... 
doi:10.6084/m9.figshare.13611278.v1 fatcat:6dso5lvowjdt3ba4y3z3gyb3l4

Estimation of multiple networks in Gaussian mixture models

Chen Gao, Yunzhang Zhu, Xiaotong Shen, Wei Pan
2016 Electronic Journal of Statistics  
A major innovation is to take advantage of the commonalities across the multiple precision matrices through possibly nonconvex fusion regularization, which for example makes it possible to achieve simultaneous  ...  Specifically, we consider penalized estimation of multiple precision matrices in the framework of a Gaussian mixture model.  ...  the fusion penalties for information borrowing across multiple cluster-specific precision matrices.  ... 
doi:10.1214/16-ejs1135 pmid:28966702 pmcid:PMC5620020 fatcat:feyrq7kgl5hjloejuoutezkhfm

Ridge Fusion in Statistical Learning [article]

Bradley S. Price, Charles J. Geyer, Adam J. Rothman
2014 arXiv   pre-print
We propose a penalized likelihood method to jointly estimate multiple precision matrices for use in quadratic discriminant analysis and model based clustering.  ...  A ridge penalty and a ridge fusion penalty are used to introduce shrinkage and promote similarity between precision matrix estimates.  ...  MODEL BASED CLUSTERING Introduction Just as in classification using QDA, semi-supervised model based clustering with Gaussian mixture models requires estimates for multiple inverse covariance matrices  ... 
arXiv:1310.3892v3 fatcat:ak5ad6b5inckvjrthxensnnt6m

Multimodal Integration: Constraining MEG Localization with EEG and fMRI [chapter]

Richard N. A. Henson
2010 International Federation for Medical and Biological Engineering Proceedings  
In the former, the addition of EEG data was shown to increase the conditional precision of source estimates relative to MEG alone; in the latter, the inclusion of each suprathreshold cluster in the fMRI  ...  I will conclude by considering whether symmetric fusion of MEG/EEG and fMRI data is worthwhile.  ...  The only way that fusion can work is if the spatial parameters estimated precisely by fMRI depend on the temporal parameters estimated precisely by MEG/EEG (or vice versa).  ... 
doi:10.1007/978-3-642-12197-5_18 fatcat:vslkd4dwgbfaxenauvkrt7hf7q

iMTF-GRN:Integrative Matrix Tri-factorization for Inference of Gene Regulatory Networks

Nisar Wani, Khalid Raza
2019 IEEE Access  
To capture more regulatory relationships with higher precision, we apply a data fusion and inference model based on Non-negative Matrix Tri-factorization called integrative matrix tri-factorization for  ...  Besides, the regulatory relationships in higher eukaryotes with large genome sizes, such as humans and mice remain mostly unexplored.  ...  in the form of kernel matrices for regularization.  ... 
doi:10.1109/access.2019.2936794 fatcat:tp3qkk3ckfcwpirqgzu6g6gx3u

3D Rigid Motion Segmentation with Mixed and Unknown Number of Models

Xun Xu, Loong Fah Cheong, Zhuwen Li
2019 IEEE Transactions on Pattern Analysis and Machine Intelligence  
From these considerations, we propose a multi-model spectral clustering framework that synergistically combines multiple models (homography and fundamental matrix) together.  ...  For general motion segmentation tasks, the number of independently moving objects is often unknown a priori and needs to be estimated from the observations.  ...  Multi-Model Spectral Clustering With multiple views provided by the different types of motion models, we have now at our disposal multiple affinity matrices.  ... 
doi:10.1109/tpami.2019.2929146 pmid:31331880 fatcat:nahqniqhine5xnijtfmb52m5ne

Covariance-Based Estimation for Clustered Sensor Networks Subject to Random Deception Attacks

Raquel Caballero-Águila, Aurora Hermoso-Carazo, Josefa Linares-Pérez
2019 Sensors  
The proposed cluster-based fusion estimation structure involves two stages.  ...  Each cluster is connected to a local processor which gathers the measured outputs of its sensors and, in turn, the local processors of all clusters are connected with a global fusion center.  ...  and precise estimators of the target signal than a single sensor.  ... 
doi:10.3390/s19143112 fatcat:3jmrbw66zzgcfn5us3offgvvny

Energy Efficient Target Tracking Method for Multi-Sensory scheduling in Wireless Sensor Networks

2020 VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE  
Experiment outcome shows proposed HF and FHF fusion model attain better performance than existing KF based method for clustered based WSN in terms of positional and velocity root mean square error and  ...  Further, to minimize the estimation errors and reduces/controlling the effects of outliers fuzzy H-infinity (FHF) filter for target tracking WSN is presented.  ...  Let represent a vector, its estimation is represented using with covariance given as .Estimation error is represented using . Two matrices and are initialized and defined by unity matrix [20] .  ... 
doi:10.35940/ijitee.c8529.019320 fatcat:t5ukxmld7vcdpbounefu25eoua

Motion Segmentation by Exploiting Complementary Geometric Models [article]

Xun Xu, Loong-Fah Cheong, Zhuwen Li
2018 arXiv   pre-print
From these considerations, we propose a multi-view spectral clustering framework that synergistically combines multiple models together.  ...  Even when we are confronted with a general scene-motion, the fundamental matrix approach as a model for motion segmentation still suffers from several defects, which we discuss in this paper.  ...  Multi-View Spectral Clustering With multiple views provided by the different types of motion models, we have now at our disposal multiple affinity matrices.  ... 
arXiv:1804.02142v1 fatcat:mm73kuejybawld574owm4acezu

Motion Segmentation by Exploiting Complementary Geometric Models

Xun Xu, Loong Fah Cheong, Zhuwen Li
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
From these considerations, we propose a multi-view spectral clustering framework that synergistically combines multiple models together.  ...  Even when we are confronted with a general scene-motion, the fundamental matrix approach as a model for motion segmentation still suffers from several defects, which we discuss in this paper.  ...  Multi-View Spectral Clustering With multiple views provided by the different types of motion models, we have now at our disposal multiple affinity matrices.  ... 
doi:10.1109/cvpr.2018.00302 dblp:conf/cvpr/XuCL18 fatcat:p3np7nrwtbhwre56ovrywxpj3y

Targeted Fused Ridge Estimation of Inverse Covariance Matrices from Multiple High-Dimensional Data Classes [article]

Anders Ellern Bilgrau, Carel F.W. Peeters, Poul Svante Eriksen, Martin Bøgsted, Wessel N. van Wieringen
2020 arXiv   pre-print
We focus on the graphical interpretation of precision matrices with the proposed estimator then serving as a basis for integrative or meta-analytic Gaussian graphical modeling.  ...  We consider the problem of jointly estimating multiple inverse covariance matrices from high-dimensional data consisting of distinct classes. An ℓ_2-penalized maximum likelihood approach is employed.  ...  A hypothesis testing literature on multiple high-dimensional precision matrices has developed concurrently with the estimation literature.  ... 
arXiv:1509.07982v2 fatcat:sztz42lvgrfxfdkitmkkhpkzja

Compressive sensing network inference with multiple-description fusion estimation

Mehdi Malboubi, Cuong Vu, Chen-Nee Chuah, Puneet Sharma
2013 2013 IEEE Global Communications Conference (GLOBECOM)  
We have previously introduced Multiple Description Fusion Estimation (MDFE) framework that partitions a largescale Under-Determined Linear Inverse (UDLI) problem into smaller sub-problems that can be solved  ...  The resulting estimates, referred to as multiple descriptions, can then be fused together to compute the global estimate [1].  ...  Therefore, NI techniques must be able to estimate these fluctuated traffic attributes with acceptable precision, depending on the application.  ... 
doi:10.1109/glocom.2013.6831295 dblp:conf/globecom/MalboubiVCS13 fatcat:5ho55ql3dbaajcbw46223dfdga

Review of Statistical Learning Methods in Integrated Omics Studies (An Integrated Information Science)

Irene Sui Lan Zeng, Thomas Lumley
2018 Bioinformatics and Biology Insights  
and covariance matrices of each cluster.  ...  Similar to the bicluster and group-regularized methods, iCluster allows faster estimation even in high-dimensional data sets.  ... 
doi:10.1177/1177932218759292 pmid:29497285 pmcid:PMC5824897 fatcat:nbknjl4qq5awrldy7natmg3h6y

Dynamic Visualization and Fast Computation for Convex Clustering via Algorithmic Regularization [article]

Michael Weylandt and John Nagorski and Genevera I. Allen
2019 arXiv   pre-print
To address these impediments, we introduce Algorithmic Regularization, an innovative technique for obtaining high-quality estimates of regularization paths using an iterative one-step approximation scheme  ...  The application of algorithmic regularization to convex clustering yields the Convex Clustering via Algorithmic Regularization Paths (CARP) algorithm for computing the clustering solution path.  ...  CARP-VIZ discards the iteration with multiple fusions, halves the step-size, and performs another iteration.  ... 
arXiv:1901.01477v3 fatcat:zt7x22tzz5ejbjm6e2x4n5m5jq
« Previous Showing results 1 — 15 out of 8,444 results