東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Projection algorithms for large scal...

Keys, Kevin Lawrence.

FindBook

Google Book

Amazon

博客來

Projection algorithms for large scale optimization and genomic data analysis.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Projection algorithms for large scale optimization and genomic data analysis./
作者:	Keys, Kevin Lawrence.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2016,
面頁冊數:	127 p.
附註:	Source: Dissertation Abstracts International, Volume: 78-03(E), Section: B.
Contained By:	Dissertation Abstracts International78-03B(E).
標題:	Mathematics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10151044
ISBN:	9781369058086

Projection algorithms for large scale optimization and genomic data analysis.
Keys, Kevin Lawrence.

Projection algorithms for large scale optimization and genomic data analysis. - Ann Arbor : ProQuest Dissertations & Theses, 2016 - 127 p.

Source: Dissertation Abstracts International, Volume: 78-03(E), Section: B.

Thesis (Ph.D.)--University of California, Los Angeles, 2016.

The advent of the Big Data era has spawned intense interest in scalable mathematical optimization methods. Traditional approaches such as Newton's method fall apart whenever the features outnumber the examples in a data set. Consequently, researchers have intensely developed first-order methods that rely only on gradients and subgradients of a cost function.

ISBN: 9781369058086Subjects--Topical Terms:

515831
Mathematics.

Projection algorithms for large scale optimization and genomic data analysis.
LDR:04034nmm a2200325 4500 001 2122455
005 20170922124917.5
008 180830s2016 ||||||||||||||||| ||eng d
020 $a 9781369058086
035 $a (MiAaPQ)AAI10151044
035 $a AAI10151044
040 $a MiAaPQ $c MiAaPQ
100 1 $a Keys, Kevin Lawrence. $3 3284428
245 1 0 $a Projection algorithms for large scale optimization and genomic data analysis.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2016
300 $a 127 p.
500 $a Source: Dissertation Abstracts International, Volume: 78-03(E), Section: B.
500 $a Adviser: Kenneth L. Lange.
502 $a Thesis (Ph.D.)--University of California, Los Angeles, 2016.
520 $a The advent of the Big Data era has spawned intense interest in scalable mathematical optimization methods. Traditional approaches such as Newton's method fall apart whenever the features outnumber the examples in a data set. Consequently, researchers have intensely developed first-order methods that rely only on gradients and subgradients of a cost function.
520 $a In this dissertation we focus on projected gradient methods for large-scale constrained optimization. We develop a particular case of a proximal gradient method called the proximal distance algorithm. Proximal distance algorithms combine the classical penalty method of constrained minimization with distance majorization. To optimize the loss function f( x) over a constraint set C, the proximal distance principle mandates minimizing the penalized loss f(x) + rho / 2 dist (x,C)2 and following the solution xrho to its limit as rho → infinity. At each iteration the squared Euclidean distance dist (x, C)2 is majorized by ||x -- pi C(xk)||2, where pi C(xk) denotes the projection of the current iterate xk onto C. The minimum of the surrogate function f(x) + rho / 2||x -- piC(x k)||2 is given by the proximal map prox rho--1f[pi C(xk)]. The next iterate xk+1 automatically decreases the original penalized loss for fixed rho. Since many explicit projections and proximal maps are known in analytic or computable form, the proximal distance algorithm provides a scalable computational framework for a variety of constraints.
520 $a For the particular case of sparse linear regression, we implement a projected gradient algorithm known as iterative hard thresholding for a particular large-scale genomics analysis known as a genome-wide association study. A genome-wide association study (GWAS) correlates marker variation with trait variation in a sample of individuals. Each study subject is genotyped at a multitude of SNPs (single nucleotide polymorphisms) spanning the genome. Here we assume that subjects are unrelated and collected at random and that trait values are normally distributed or transformed to normality. Over the past decade, researchers have been remarkably successful in applying GWAS analysis to hundreds of traits. The massive amount of data produced in these studies present unique computational challenges. Penalized regression with LASSO or MCP penalties is capable of selecting a handful of associated SNPs from millions of potential SNPs. Unfortunately, model selection can be corrupted by false positives and false negatives, obscuring the genetic underpinning of a trait. Our parallel implementation of IHT accommodates SNP genotype compression and exploits multiple CPU cores and graphics processing units (GPUs). This allows statistical geneticists to leverage desktop workstations in GWAS analysis and to eschew expensive supercomputing resources. We evaluate IHT performance on both simulated and real GWAS data and conclude that it reduces false positive and false negative rates while remaining competitive in computational time with penalized regression.
590 $a School code: 0031.
650 4 $a Mathematics. $3 515831
650 4 $a Bioinformatics. $3 553671
650 4 $a Biostatistics. $3 1002712
690 $a 0405
690 $a 0715
690 $a 0308
710 2 $a University of California, Los Angeles. $b Biomathematics 0121. $3 3284429
773 0 $t Dissertation Abstracts International $g 78-03B(E).
790 $a 0031
791 $a Ph.D.
792 $a 2016
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10151044