Filters








11,895 Hits in 6.3 sec

A framework to characterize performance of LASSO algorithms [article]

Mihailo Stojnic
2013 arXiv   pre-print
Moreover, as a by-product of our analysis framework we recognize existence of an SOCP type of algorithm that achieves the same performance.  ...  unknown vector (of sparsity proportional to the length of the vector) can be recovered from an under-determined system via a simple polynomial ℓ_1-optimization algorithm.  ...  In this paper we will develop a novel framework for performance characterization of the LASSO algorithms.  ... 
arXiv:1303.7291v1 fatcat:hq3xlpjjyvc3javmxyuhbncndy

High-Dimensional Screening Using Multiple Grouping of Variables

Divyanshu Vats
2014 IEEE Transactions on Signal Processing  
We propose a novel framework for screening, which we refer to as Multiple Grouping (MuG), that groups variables, performs variable selection over the groups, and repeats this process multiple number of  ...  times to estimate a sequence of sets that contains the non-zero entries in \beta*.  ...  We study the application of MuG with group Lasso [5] , which uses a modification of the popular Lasso algorithm [6] to perform variable selection over groups.  ... 
doi:10.1109/tsp.2013.2294591 fatcat:6wnikgmdrzgybd562rkoblqy4i

PyUoI: The Union of Intersections Framework in Python

Pratik Sachdeva, Jesse Livezey, Andrew Tritt, Kristofer Bouchard
2019 Journal of Open Source Software  
Acknowledgements We thank the contributors to earlier versions of this software.  ...  P.S.S. was supported by the Department of Defense (DoD) through the National Defense Science & Engineering Graduate Fellowship (NDSEG) Program.  ...  Background The Union of Intersections is not a single method or algorithm, but a flexible statistical framework into which other algorithms can be inserted.  ... 
doi:10.21105/joss.01799 fatcat:7f57nkuvfrgi5b3swj7yh37k3y

Workload Characterization at the Virtualization Layer

Fatemeh Azmandian, Micha Moffie, Jennifer G. Dy, Javed A. Aslam, David R. Kaeli
2011 2011 IEEE 19th Annual International Symposium on Modelling, Analysis, and Simulation of Computer and Telecommunication Systems  
In this paper, we present a novel approach to working at a virtualization interface, performing workload characterization equipped with the information available at the virtual machine monitor (VMM) interface  ...  We show that we are able to capture enough information about a workload to characterize and decompose it into a combination of CPU, memory, disk I/O, and network I/O-intensive components.  ...  Regression algorithms have been previously applied to characterize the performance of workloads.  ... 
doi:10.1109/mascots.2011.63 dblp:conf/mascots/AzmandianMDAK11 fatcat:bgm7braklbawti6mazekqekdam

An integrated feature ranking and selection framework for ADHD characterization

Cao Xiao, Jesse Bledsoe, Shouyi Wang, Wanpracha Art Chaovalitwongse, Sonya Mehta, Margaret Semrud-Clikeman, Thomas Grabowski
2016 Brain Informatics  
The proposed framework combines information theoretic criteria and the least absolute shrinkage and selection operator (Lasso) method into a two-step feature selection process which is capable of selecting  ...  To tackle this problem, this paper proposes a novel integrated feature ranking and selection framework that utilizes normalized brain cortical thickness features extracted from MRI data to discriminate  ...  Performance evaluation using simulated dataset To evaluate the performance of the proposed feature selection framework, we used a simulated dataset with binary response and contain p ¼ 45 predictors and  ... 
doi:10.1007/s40708-016-0047-1 pmid:27747592 pmcid:PMC4999568 fatcat:ro4jyx3gdrd5tby3d6fikb4kou

Chunk Dynamic Updating for Group Lasso with ODEs

Diyang Li, Bin Gu
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Based on the new formulation, we propose a path following algorithm for group Lasso regarding to the adjustment parameter.  ...  However, most of existing algorithms of group Lasso are limited to offline updating, and only one is online algorithm which can only handle newly added samples inexactly.  ...  To prove the practicability of Algorithm 1, we compare the training process of CIGL and CDGL with batch algorithm and existing online framework for group lasso using dual averaging method, i.e., DA-GL  ... 
doi:10.1609/aaai.v36i7.20704 fatcat:slvl2wxtcfdxznzekg3w3sv3ju

Exploring stability-based voxel selection methods in MVPA using cognitive neuroimaging data: a comprehensive study

Miaolin Fan, Chun-An Chou
2016 Brain Informatics  
to a bootstrapped stability selection framework, and the selected algorithms were compared based on both accuracy and stability scores.  ...  The results indicate that regularizationbased methods are generally more stable in StarPlus dataset, but in Haxby dataset they failed to perform as well as others.  ...  The basic concept is to characterize BOLD signals by fitting GLM to a haemodynamic response function (HRF) that describes blood-oxygen-level responses to the given stimulus as a function of time.  ... 
doi:10.1007/s40708-016-0048-0 pmid:27747593 pmcid:PMC4999569 fatcat:dyn5e6xscna5xoqpjwuz3nmqxe

Clustered Compressed Sensing in fMRI Data Analysis Using a Bayesian Framework

Solomon Tesfamicael
2014 International Journal of Information and Electronics Engineering  
Index Terms-Bayesian framework, sparse prior, clustered prior, posterior, MAP, compressive sensing, LASSO, clustered LASSO, GLM, fMRI data.  ...  This enhances the effectiveness of the model to help analyze the data better. So in this work we have built the Bayesian framework needed first.  ...  And we used most of the algorithms mentioned above to compare their performance.  ... 
doi:10.7763/ijiee.2014.v4.412 fatcat:f3nc4e37lngc5e3vkg33w6djpq

A framework of irregularity enlightenment for data pre-processing in data mining

Siu-Tong Au, Rong Duan, Siamak G. Hesar, Wei Jiang
2008 Annals of Operations Research  
This paper proposes a generic Irregularity Enlightenment (IE) framework for dealing with the situation when multiple irregularities are hidden in large volumes of data in general and cross sectional time  ...  By decomposing time series data into basic components, we propose to optimize a penalized least square loss function to aid the selection of key irregularities in consecutive steps and cluster time series  ...  Acknowledgements The authors would like to thank the editors and two anonymous referees for their valuable suggestions, which have significantly improved the quality and presentation of this paper.  ... 
doi:10.1007/s10479-008-0494-z fatcat:mwyz7fv2tvgcho45y2qelofqye

Penalized regression procedures for variable selection in the potential outcomes framework

Debashis Ghosh, Yeying Zhu, Donna L. Coffman
2015 Statistics in Medicine  
The framework leads to a simple 'impute, then select' class of procedures that is agnostic to the type of imputation algorithm as well as penalized regression used.  ...  A shared LASSO and difference LASSO algorithm are defined, along with their multiple imputation analogues. The procedures are illustrated using a well-known right heart catheterization dataset.  ...  Thus, we wish to perform a LASSO of the multivariate outcome on covariates. Thus, its use is intimately tied to the type of variable selection we wish to perform.  ... 
doi:10.1002/sim.6433 pmid:25628185 pmcid:PMC4390482 fatcat:ohmitt4a4rgntfljtarci64ose

FEAST: An Automated Feature Selection Framework for Compilation Tasks [article]

Pai-Shun Ting, Chun-Chen Tu, Pin-Yu Chen, Ya-Yun Lo, Shin-Ming Cheng
2016 arXiv   pre-print
The success of the application of machine-learning techniques to compilation tasks can be largely attributed to the recent development and advancement of program characterization, a process that numerically  ...  While great achievements have been made in identifying key features to characterize programs, choosing a correct set of features for a specific compiler task remains an ad hoc procedure.  ...  In their work, an SVM-based supervised training algorithm is used to train a set of support vectors that can help estimate the performance or reaction of an unknown program to a set of compiler parameters  ... 
arXiv:1610.09543v1 fatcat:movljxz7k5h2blak3nnwuouv3m

A Novel Graphical Lasso based approach towards Segmentation Analysis in Energy Game-Theoretic Frameworks [article]

Hari Prasanna Das, Ioannis C. Konstantakopoulos, Aummul Baneen Manasawala, Tanya Veeravalli, Huihan Liu, Costas J. Spanos
2019 arXiv   pre-print
We propose a novel graphical lasso based approach to perform such segmentation, by studying the feature correlations in a real-world energy social game dataset.  ...  Our research proposes that the utilities of players in such a framework can be grouped together to a relatively small number of clusters, and the clusters can then be targeted with tailored incentives.  ...  Knitting together both the methods via a powerful tool, the graphical lasso algorithm, we present a novel methodology to perform segmentation in energy game-theoretic frameworks.  ... 
arXiv:1910.02217v1 fatcat:5qf453pujzd27imme76p67p474

A Bayesian Lasso via reversible-jump MCMC

Xiaohui Chen, Z. Jane Wang, Martin J. McKeown
2011 Signal Processing  
As an extension of the proposed RJ-MCMC framework, we also develop an MCMC-based algorithm for the Binomial-Gaussian prior model and illustrate its improved performance over the non-Bayesian estimate via  ...  To obtain the Bayesian Lasso estimate, a reversiblejump MCMC algorithm is developed for joint posterior inference over both discrete and continuous parameter spaces.  ...  This work was supported by a Pacific Alzheimer's Research Foundation (PARF) Centre Grant Award.  ... 
doi:10.1016/j.sigpro.2011.02.014 fatcat:fx3osa2u4zcrljs76r3axwzw6u

Exact Post Model Selection Inference for Marginal Screening [article]

Jason D Lee, Jonathan E Taylor
2014 arXiv   pre-print
At the core of this framework is a result that characterizes the exact distribution of linear functions of the response y, conditional on the model being selected ("condition on selection" framework).  ...  We show how to apply the proposed framework to several other selection procedures including orthogonal matching pursuit, non-negative least squares, and marginal screening+Lasso.  ...  Jason Lee was supported by a NSF graduate fellowship, and a Stanford Graduate Fellowship.  ... 
arXiv:1402.5596v2 fatcat:xfhhnotfg5g2bk754datbn5hay

Asymptotic Analysis of Complex LASSO via Complex Approximate Message Passing (CAMP) [article]

Arian Maleki, Laura Anitori, Zai Yang, Richard Baraniuk
2013 arXiv   pre-print
While several studies have shown that the LASSO algorithm offers desirable solutions under certain conditions, the precise asymptotic performance of this algorithm in the complex setting is not yet known  ...  We then generalize the state evolution framework recently introduced for the analysis of AMP, to the complex setting.  ...  Formal analysis of CAMP and c-LASSO In this section, we explain the state evolution framework as a framework that predicts the performance of the CAMP and c-LASSO in the asymptotic settings.  ... 
arXiv:1108.0477v2 fatcat:btptzuh7fvgkdjyxsb72yi6xdy
« Previous Showing results 1 — 15 out of 11,895 results