Cross-validation prior choice in Bayesian probit regression with many covariates

Lamnisos, Demetris; Griffin, Jim E.; Steel, Mark F. J.

doi:10.1007/s11222-011-9228-1

Cited by 12 publications

(9 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification was accomplished with Bayesian logistic regression with a sparsity promoting Laplace prior (see [ 32 , 33 ] for mathematical description of prior). Each individual voxel weight within a ROI was given a univariate Laplace prior distribution with a scale hyperparameter, which was optimized separately for each subject or subject pair by maximizing the average accuracy over all other subjects or subject pairs ([ 34 ]; candidate values 0.01, 0.04, 0.21, 1, 4.64, 21.54, 100). The multivariate posterior distribution of classifier weights was approximated using the expectation propagation algorithm [ 33 ] implemented in the FieldTrip toolbox [ 35 ].…”

Section: Methodsmentioning

confidence: 99%

Brain-to-brain hyperclassification reveals action-specific motor mapping of observed actions in humans

et al. 2017

View full text Add to dashboard Cite

Seeing an action may activate the corresponding action motor code in the observer. It remains unresolved whether seeing and performing an action activates similar action-specific motor codes in the observer and the actor. We used novel hyperclassification approach to reveal shared brain activation signatures of action execution and observation in interacting human subjects. In the first experiment, two "actors" performed four types of hand actions while their haemodynamic brain activations were measured with 3-T functional magnetic resonance imaging (fMRI). The actions were videotaped and shown to 15 "observers" during a second fMRI experiment. Eleven observers saw the videos of one actor, and the remaining four observers saw the videos of the other actor. In a control fMRI experiment, one of the actors performed actions with closed eyes, and five new observers viewed these actions. Bayesian canonical correlation analysis was applied to functionally realign observers' and actors' fMRI data. Hyperclassification of the seen actions was performed with Bayesian logistic regression trained on actors' data and tested with observers' data. Without the functional realignment, between-subjects accuracy was at chance level. With the realignment, the accuracy increased on average by 15 percentage points, exceeding both the chance level and the accuracy without functional realignment. The highest accuracies were observed in occipital, parietal and premotor cortices. Hyperclassification exceeded chance level also when the actor did not see her own actions. We conclude that the functional brain activation signatures underlying action execution and observation are partly shared, yet these activation signatures may be anatomically misaligned across individuals.

show abstract

Section: Methodsmentioning

confidence: 99%

Brain-to-brain hyperclassification reveals action-specific motor mapping of observed actions in humans

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Our MVPA analysis was based on Bayesian treatment of logistic regression classifiers that attach a given transient activation pattern to the more probable one of two stimulus classes ( For the final models trained using the data of all 16 participants, the scale hyperparameter of the Laplace prior was optimized (candidate values 10 k O , where { 6, 5, 4, 3, 2} k ) by maximizing the mean log predictive probability (MLPP) obtained in a leave-one-out crossvalidation across participants (one participant at a time was left out from the training data set and the model was trained using the remaining 15 participants) and averaged over all seven binary classification tasks (Lamnisos et al 2012). The posterior distribution of the voxel coefficients was visualized by presenting the marginal posterior probabilities for positive sign of each coefficient as a brain map, which we call a signature pattern.…”

Section: Multi-voxel Pattern Analysismentioning

confidence: 99%

Distributed neural signatures of natural audiovisual speech and music in the human auditory cortex

et al. 2017

View full text Add to dashboard Cite

Distributed neural signatures of natural audiovisual speech and music in the human auditory cortex, NeuroImage, http://dx.doi.org/10. 1016/j.neuroimage.2016.12.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Graphical abstract

show abstract

“…Let

be the binary outcome vector in the partition

, and

be the outcome from the remaining partitions (training set). The cross-validation density for the i -th individual is estimated via model averaging; following Gelfand and Dey, 1994 , Lamnisos et al, 2012 , Chekouo et al, 2017 , we use

as the importance density of the target distribution

, where

. The predictive probability can then be written as

where

, and G is the number of distinct models

’s obtained from the MCMC samples.…”

Section: Posterior Inferencementioning

confidence: 99%

A Bayesian 2D functional linear model for gray-level co-occurrence matrices in texture analysis of lower grade gliomas

Chekouo

Mohammed

Rao

2020

NeuroImage: Clinical

View full text Add to dashboard Cite

Highlights We treat gray-level co-occurrence matrices (GLCM) as two-dimensional functional data objects. We develop a Bayesian functional regression method that relates GLCMs to IDH mutation status. The method accounts for multiple imaging sequences. The method outperforms other competing methods that use only GLCM derived summary statistics.

show abstract

Cross-validation prior choice in Bayesian probit regression with many covariates

Cited by 12 publications

References 28 publications

Brain-to-brain hyperclassification reveals action-specific motor mapping of observed actions in humans

Brain-to-brain hyperclassification reveals action-specific motor mapping of observed actions in humans

Distributed neural signatures of natural audiovisual speech and music in the human auditory cortex

A Bayesian 2D functional linear model for gray-level co-occurrence matrices in texture analysis of lower grade gliomas

Contact Info

Product

Resources

About