10,356 Hits in 8.2 sec

Learning High-dimensional Gaussian Graphical Models under Total Positivity without Adjustment of Tuning Parameters [article]

Yuhao Wang, Uma Roy, Caroline Uhler
2020 arXiv   pre-print
We here propose a new method to estimate the underlying undirected graphical model under MTP2 and show that it is provably consistent in structure recovery without adjusting the tuning parameters.  ...  We consider the problem of estimating an undirected Gaussian graphical model when the underlying distribution is multivariate totally positive of order 2 (MTP2), a strong form of positive dependence.  ...  At the time this research was completed, Yuhao Wang and Uma Roy was at the Massachusetts Institute of Technology.  ... 
arXiv:1906.05159v4 fatcat:tjz7djpumvcftdvd2k7knxscea

A sparse conditional Gaussian graphical model for analysis of genetical genomics data

Jianxin Yin, Hongzhe Li
2011 Annals of Applied Statistics  
In this paper we introduce a sparse conditional Gaussian graphical model for studying the conditional independent relationships among a set of gene expressions adjusting for possible genetic effects where  ...  We apply our methods to the analysis of a yeast eQTL data set and demonstrate that the conditional Gaussian graphical model leads to a more interpretable gene network than a standard Gaussian graphical  ...  Acknowledgments We thank the three reviewers and the editor for many insightful comments that have greatly improved the presentation of this paper.  ... 
doi:10.1214/11-aoas494 pmid:22905077 pmcid:PMC3419502 fatcat:7u47p24efvfm5ngcz5itcx4o2m

Model-based clustering in very high dimensions via adaptive projections [article]

Bernd Taschler, Frank Dondelinger, Sach Mukherjee
2019 arXiv   pre-print
However, when the dimension p is large relative to sample size n and where either or both of means and covariances/graphical models may differ between the latent groups, mixture models face statistical  ...  Combining a full covariance formulation with the adaptive projection allows detection of both mean and covariance signals in very high dimensional problems.  ...  High-dimensional graphical model estimation In the examples above we focused on assignment risk in high-dimensional mixture modelling but did not directly discuss estimation of the high-dimensional parameters  ... 
arXiv:1902.08472v1 fatcat:4thnbchjrfbkbazl6ka3wfswta

A Simple Correction Procedure for High-Dimensional Generalized Linear Models with Measurement Error [article]

Michael Byrd, Monnie McGee
2020 arXiv   pre-print
We consider high-dimensional generalized linear models when the covariates are contaminated by measurement error.  ...  We apply our correction to gene microarray data, and illustrate that it results in a great reduction in the number of false positives whilst still retaining most true positives.  ...  We consider the high-dimensional EIV regression problem for GLMs with response y ∼ D and nuissance parameters Θ.  ... 
arXiv:1912.11740v2 fatcat:n7hjs2fe5bcwdf42lrjftkcidi

A Lightweight CNN Based on Transfer Learning for COVID-19 Diagnosis

Xiaorui Zhang, Jie Zhou, Wei Sun, Sunil Kumar Jha
2022 Computers Materials & Continua  
In order to alleviate the problem of model overfitting caused by insufficient data set, transfer learning is used to train the model.  ...  computers without GPU acceleration.  ...  the models without use transfer learning.  ... 
doi:10.32604/cmc.2022.024589 fatcat:zbcfcaf6kjhu7ine2yzqhze7yq

Learning complex dependency structure of gene regulatory networks from high dimensional micro-array data with Gaussian Bayesian networks [article]

Catharina Elisabeth Graafland, José Manuel Gutiérrez
2022 arXiv   pre-print
In the Undirected probabilistic Graphical Model (UGM) framework the Glasso algorithm has been proposed to deal with high dimensional micro-array datasets forcing sparsity.  ...  In this work we advocate the use of a simple score-based Hill Climbing algorithm (HC) that learns Gaussian Bayesian Networks (BNs) leaning on Directed Acyclic Graphs (DAGs).  ...  Acknowledgements CEG would like to acknowledge the support of the funding from the Spanish Agencia Estatal de Investigación through the Unidad de Excelencia María de Maeztu with reference MDM-2017-0765  ... 
arXiv:2106.15365v2 fatcat:hyv5ygjrxzewhbmay5zfqn4d3m

Covariate-adjusted precision matrix estimation with an application in genetical genomics

T. T. Cai, H. Li, W. Liu, J. Xie
2012 Biometrika  
Motivated by analysis of genetical genomics data, we introduce a sparse high-dimensional multivariate regression model for studying conditional independence relationships among a set of genes adjusting  ...  Simulation shows that the proposed method results in significant improvements in both precision matrix estimation and graphical structure selection when compared to the standard Gaussian graphical model  ...  ACKNOWLEDGEMENT This research was supported by the National Institutes of Health, the National Science Foundation, the Program for Professor of Special Appointment at Shanghai Institutions of Higher Learning  ... 
doi:10.1093/biomet/ass058 pmid:28316337 pmcid:PMC5351557 fatcat:viitg5m7a5ac7g67zvhjqu3pya

Robust causal structure learning with some hidden variables

Benjamin Frot, Preetam Nandy, Marloes H. Maathuis
2019 Journal of The Royal Statistical Society Series B-statistical Methodology  
This approach is consistent in certain high dimensional regimes and performs favourably when compared with the state of the art, in terms of both graphical structure recovery and total causal effect estimation  ...  DAG under the assumption that there are no remaining hidden variables.  ...  We are also indebted to the reviewers for their suggestions on how to expand the scope of the paper and for pointing out some inconsistencies.  ... 
doi:10.1111/rssb.12315 fatcat:5e2nil3jxzft3nlzjpeonyscxu

A Completely Tuning-Free and Robust Approach to Sparse Precision Matrix Estimation

Chau Tran, Guo Yu
2022 International Conference on Machine Learning  
We propose a completely tuning-free approach for estimating sparse Gaussian graphical models.  ...  Despite the vast literature on sparse Gaussian graphical models, current methods either are asymptotically tuning-free (which still require fine-tuning in practice) or hinge on computationally expensive  ...  Contributions: In this paper, we propose a completely tuning-free method in high-dimensional Gaussian graphical models.  ... 
dblp:conf/icml/TranY22 fatcat:zi7qxygy7nevzo6hmvcufu2aka

Robust causal structure learning with some hidden variables [article]

Benjamin Frot, Preetam Nandy, Marloes H. Maathuis
2018 arXiv   pre-print
This approach is consistent in certain high-dimensional regimes and performs favourably when compared to the state of the art, both in terms of graphical structure recovery and total causal effect estimation  ...  underlying DAG under the assumption that there are no remaining hidden variables.  ...  We are also indebted to the anonymous reviewers for their suggestions on how to expand to the scope of our paper and for pointing out some inconsistencies.  ... 
arXiv:1708.01151v2 fatcat:nmrap5u33fbavjd2yn7ktxshom

Apple flower detection using deep convolutional networks

Philipe A. Dias, Amy Tabb, Henry Medeiros
2018 Computers in industry (Print)  
is fine-tuned to become specially sensitive to flowers.  ...  Moreover, a performance assessment on three additional datasets previously unseen by the network, which consist of different flower species and were acquired under different conditions, reveals that the  ...  Two main parameters control the performance of SVMs with a Gaussian kernel function, the regularization cost C and the width of the Gaussian kernel γ.  ... 
doi:10.1016/j.compind.2018.03.010 fatcat:x25y63ud4jeoff7q4xlkiyfpda

Molecular heterogeneity at the network level: high-dimensional testing, clustering and a TCGA case study

Nicolas Städler, Frank Dondelinger, Steven M Hill, Rehan Akbani, Yiling Lu, Gordon B Mills, Sach Mukherjee, Cenk Sahinalp
2017 Bioinformatics  
Results: We leverage recent ideas from high-dimensional statistics for testing and clustering in the network biology setting.  ...  Furthermore, since network differences could provide important and biologically interpretable information to identify molecular subgroups, there is a need to consider the unsupervised task of learning  ...  Subtype identification using MixGlasso MixGlasso is a penalized mixture of Gaussian graphical models.  ... 
doi:10.1093/bioinformatics/btx322 pmid:28535188 pmcid:PMC5590725 fatcat:w22w4hehfvhcdhax2p3qeyv6mm

Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection

Bo Zong, Qi Song, Martin Renqiang Min, Wei Cheng, Cristian Lumezanu, Dae-ki Cho, Haifeng Chen
2018 International Conference on Learning Representations  
in an end-to-end fashion, leveraging a separate estimation network to facilitate the parameter learning of the mixture model.  ...  Unsupervised anomaly detection on multi-or high-dimensional data is of great importance in both fundamental machine learning research and industrial applications, for which density estimation lies at the  ...  detection, and the estimation network evaluates sample energy in the low-dimensional space under the framework of Gaussian Mixture Modeling.  ... 
dblp:conf/iclr/ZongSMCLCC18 fatcat:3o3osxs7k5epvmgwnuwgpbcvn4

Statistical Approaches for the Study of Cognitive and Brain Aging

Huaihou Chen, Bingxin Zhao, Guanqun Cao, Eric C. Proges, Andrew O'Shea, Adam J. Woods, Ronald A. Cohen
2016 Frontiers in Aging Neuroscience  
potential nonlinear age effects. (2) Marginal correlations are often used in brain network analysis, which are not efficient in characterizing a complex brain network. (3) Due to the challenge of high-dimensionality  ...  Specifically, we introduce semiparametric models for modeling age effects, graphical models for brain network analysis, and penalized regression methods for selecting the most important markers in predicting  ...  However, the high-dimensionality of the neuroimaging markers posit challenges on how to efficiently pick up the informative subset of the markers.  ... 
doi:10.3389/fnagi.2016.00176 pmid:27486400 pmcid:PMC4949247 fatcat:hehbjsp7h5b4bftaumhkebwree

Microbial Networks in SPRING - Semi-parametric Rank-Based Correlation and Partial Correlation Estimation for Quantitative Microbiome Data

Grace Yoon, Irina Gaynanova, Christian L. Müller
2019 Frontiers in Genetics  
Combining this estimator with sparse graphical modeling techniques leads to the Semi-Parametric Rank-based approach for INference in Graphical model (SPRING).  ...  SPRING shows superior network recovery performance on a wide range of realistic benchmark problems with varying network topologies and is robust to misspecifications of the total cell count estimate.  ...  To construct graphical models for the truncated Gaussian copula model, the estimation of the latent correlation matrix is required.  ... 
doi:10.3389/fgene.2019.00516 pmid:31244881 pmcid:PMC6563871 fatcat:jn45atosgvaidb773vmlhgm7ia
« Previous Showing results 1 — 15 out of 10,356 results