Filters








1,405 Hits in 4.1 sec

A whitening approach to probabilistic canonical correlation analysis for omics data integration [article]

Takoua Jendoubi, Korbinian Strimmer
2018 arXiv   pre-print
The advantages of this variant of probabilistic CCA include non-ambiguity of the latent variables, provisions for negative canonical correlations, possibility of non-normal generative variables, as well  ...  Canonical correlation analysis (CCA) is a classic statistical tool for investigating complex multivariate data.  ...  Acknowledgements KS thanks Martin Lotz for discussion. TJ was funded by a Wellcome Trust ISSF Ph.D. studentship. We also thank the anonymous referees for their many useful suggestions.  ... 
arXiv:1802.03490v3 fatcat:dn22d3nlgbhajhoyekase36bda

A Survey of Multi-View Representation Learning [article]

Yingming Li, Ming Yang, Zhongfei Zhang
2017 arXiv   pre-print
Representative examples are canonical correlation analysis (CCA) and its several extensions.  ...  , and multi-view latent space Markov networks, to neural network-based methods including multi-modal autoencoders, multi-view convolutional neural networks, and multi-modal recurrent neural networks.  ...  Probabilistic CCA CCA can be interpreted as the maximum likelihood solution to a probabilistic latent variable model for two Gaussian random vectors.  ... 
arXiv:1610.01206v4 fatcat:xsi7ufxnlbdk5lz6ykrsnexfvm

Probabilistic penalized principal component analysis

Chongsun Park, Morgan C. Wang, Eun Bi Mo
2017 Communications for Statistical Applications and Methods  
A variable selection method based on probabilistic principal component analysis (PCA) using penalized likelihood method is proposed. The proposed method is a two-step variable reduction method.  ...  It is straightforward to extend our likelihood method in handling problems with missing observations using EM algorithms.  ...  All numerical comparisons are conducted using R codes.  ... 
doi:10.5351/csam.2017.24.2.143 fatcat:dhvcmfj2ujbsfeqhvwbygjf33m

A combined latent class and trait model for the analysis and visualization of discrete data

A. Kaban, M. Girolami
2001 IEEE Transactions on Pattern Analysis and Machine Intelligence  
AbstractÐWe present a general framework for data analysis and visualization by means of topographic organization and clustering.  ...  The combined (trait and class) model along with the associated estimation procedures allows us to interpret the visualization result, in the sense of a topographic ordering.  ...  Some of the routines of the MATLAB GTM Toolbox http://www.ncrg.aston.ac.uk/GTM were used in the software for the reported simulations. The RainBow toolkit http://www.cs.cmu.edu/mccallum/bow/  ... 
doi:10.1109/34.946989 fatcat:nkfkzduxpzb7rpfgyay3jvqpdq

Similarity Gaussian Process Latent Variable Model for Multi-modal Data Analysis

Guoli Song, Shuhui Wang, Qingming Huang, Qi Tian
2015 2015 IEEE International Conference on Computer Vision (ICCV)  
They can be applied to various tasks to discover the non-linear correlations and obtain the comparable low-dimensional representation for heterogeneous modalities.  ...  On two widely used real-world datasets, we outperform previous approaches for cross-modal content retrieval and cross-modal classification.  ...  As a generative model, MLBE uses binary hash codes as latent variables to generate intra-modal and inter-modal similarities. The code length for MLBE is set to 8 in our experiments.  ... 
doi:10.1109/iccv.2015.461 dblp:conf/iccv/SongWHT15 fatcat:zzaphoi2knc7zgcvwiuipmisna

A new stochastic multidimensional unfolding model for the investigation of paired comparison consumer preference/choice data

Wayne S. DeSarbo, Geert De Soete, Jehoshua Eliashberg
1987 Journal of Economic Psychology  
To illustrate tl e versatility of the model, a small application measuring consumer preference for severa!  ...  The canonical correlations are lowest for the ideal points.  ...  According to Anderson (1980) , maximum likelihood estimators in such cases may not be consistent.  ... 
doi:10.1016/0167-4870(87)90029-8 fatcat:ajeiyy2isjdp5cqtw6grtfdxoy

A Probabilistic Multidimensional Scaling Vector Model

Wayne S. DeSarbo, Richard L. Oliver, Geen De Soete
1986 Applied Psychological Measurement  
A small monte carlo analysis performed on synthetic data with the new method is also presented.  ...  and stimuli as points in a Tdimensional space, where the scalar products, or projections of the stimulus points onto the subject vectors, provide respective information as to the utility (or whatever latent  ...  &dquo; Then, Maximum likelihood procedures are used to estimate A = lIa¡¡11 and B = Ilbitll, given A = Ilô¡ikll and T.  ... 
doi:10.1177/014662168601000107 fatcat:anlvgbi2dva63osi6fs3upz57u

Multimodal Data Fusion in High-Dimensional Heterogeneous Datasets via Generative Models [article]

Yasin Yilmaz, Mehmet Aktukmak, Alfred O. Hero
2021 arXiv   pre-print
The commonly used latent space embedding techniques, such as Principal Component Analysis, Factor Analysis, and manifold learning techniques, are typically used for learning effective representations of  ...  In this paper, we are interested in learning probabilistic generative models from high-dimensional heterogeneous data in an unsupervised fashion.  ...  Probabilistic Canonical Correlation Analysis (PCCA) [12] similarly models a pair of Gaussian feature vectors using a weighted sum of latent factors, called canonical components.  ... 
arXiv:2108.12445v2 fatcat:25fbkkk7h5ebjmublgxiwy536q

Probabilistic analysis of the human transcriptome with side information [article]

Leo Lahti
2011 arXiv   pre-print
Statistical learning and probabilistic models provide a natural framework for such modeling tasks.  ...  The key contributions of the thesis are general exploratory tools for high-throughput data analysis that have provided new insights to cell-biological networks, cancer mechanisms and other aspects of genome  ...  Classical and probabilistic canonical correlation analysis Canonical correlation analysis (CCA) is a classical method for detecting linear dependencies between two multivariate random variables (Hotelling  ... 
arXiv:1102.5509v1 fatcat:zlwxrob7c5hapgje2v2tstx6uy

Motion Models for People Tracking [chapter]

David J. Fleet
2011 Visual Analysis of Humans  
This chapter provides an introduction to models of human pose and motion for use in 3D human pose tracking.  ...  We concentrate on probabilistic latent variable models of kinematics, most of which are learned from motion capture data, and on recent physics-based models.  ...  It can be used as a probabilistic dimensionality reduction, where the latent variables capture the structure (latent causes) of the high-dimensional training data.  ... 
doi:10.1007/978-0-85729-997-0_10 fatcat:fr7d7dpvhfepvdwh4mkhfpcqby

Stimulus-choice (mis)alignment in primate MT cortex [article]

Yuan Zhao, Jacob Lachenmyer Yates, Aaron Joseph Levi, Alexander Christopher Huk, Il Memming Park
2019 bioRxiv   pre-print
We dissected the particular contributions of sensory-driven versus choice-correlated activity in the low-dimensional population code.  ...  Using a statistical nonlinear dimensionality reduction technique on single-trial ensemble recordings from the middle temporal area during perceptual-decision-making, we extracted low-dimensional neural  ...  Fig 2 . 2 Probabilistic description of a single trial using variational latent Gaussian process method and resulting noise correlation.  ... 
doi:10.1101/2019.12.20.884189 fatcat:fh2avtm3jncbdddw6qmvdrs2li

Stimulus-choice (mis)alignment in primate area MT

Yuan Zhao, Jacob L. Yates, Aaron J. Levi, Alexander C. Huk, Il Memming Park, Daniele Marinazzo
2020 PLoS Computational Biology  
Using a statistical nonlinear dimensionality reduction technique on single-trial ensemble recordings from the middle temporal (MT) area during perceptual-decision-making, we extracted low-dimensional latent  ...  We dissected the particular contributions of sensory-driven versus choice-correlated activity in the low-dimensional population code.  ...  Acknowledgments We thank the anonymous reviewers for their helpful comments. Memming thanks Hendrikje Nienborg for stimulating discussions.  ... 
doi:10.1371/journal.pcbi.1007614 pmid:32421716 fatcat:mjfgobg3trfc3ddvyzwutjohky

Probabilistic Independent Component Analysis for Functional Magnetic Resonance Imaging

C.F. Beckmann, S.M. Smith
2004 IEEE Transactions on Medical Imaging  
In order to avoid overfitting, we employ objective estimation of the amount of Gaussian noise through Bayesian analysis of the true dimensionality of the data, i.e. the number of activation and non-Gaussian  ...  We present an integrated approach to Probabilistic ICA for FMRI data that allows for nonsquare mixing in the presence of Gaussian noise.  ...  Dave McGonigle for providing parts of the data used within this paper.  ... 
doi:10.1109/tmi.2003.822821 pmid:14964560 fatcat:oyyr6qqjy5hfrglsdelap6ynwq

Tensor canonical correlation analysis

Eun Jeong Min, Eric C. Chi, Hua Zhou
2019 Stat  
Canonical correlation analysis (CCA) is a multivariate analysis technique for estimating a linear relationship between two sets of measurements.  ...  Also we describe a probabilistic model for TCCA that enables the generation of synthetic data with desired canonical variates and correlations.  ...  The maximum likelihood estimation may improve upon the above estimates when data actually come from an array normal distribution.  ... 
doi:10.1002/sta4.253 pmid:32655193 pmcid:PMC7351364 fatcat:myoar726lfg67fhi7ypqze6uyq

On Independent Component Analysis for Multimedia Signals [chapter]

Thomas Kolenda, Jan Larsen, Lars Kai Hansen
2000 Multimedia Image and Video Processing  
The objective of this paper is to demonstrate the potential of independent component analysis and blind sources separation methods for modeling and understanding of multimedia data, which largely refers  ...  Finally, we provide a detailed presentation of our own recent work on modeling combined text/image data for the purpose of cross-media retrieval.  ...  Joint Probabilistic Latent Semantic Analysis (PLSA) is used to identify a common latent space for predicting terms and links.  ... 
doi:10.1201/9781420037562.ch7 fatcat:weavyd2fj5hixj4zl65jatzqdy
« Previous Showing results 1 — 15 out of 1,405 results