國立高雄大學圖資館 |

Language: English

Back

Bayesian Additive Regression Networks (BARN): A Machine Learning Algorithm and Scientific Software Library /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Bayesian Additive Regression Networks (BARN): A Machine Learning Algorithm and Scientific Software Library /Danielle Van Boxel.
Author:	Van Boxel, Danielle,
Description:	1 electronic resource (111 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
Notes:	Advisors: Roman-Palacios, Cristian; Tang, Xueying Committee members: Maccabe, Arthur Barney; Zhang, Hao Helen; Kennedy, Tom.
Contained By:	Dissertations Abstracts International85-11B.
Subject:	Applied mathematics.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31241398
ISBN:	9798382322223

Bayesian Additive Regression Networks (BARN): A Machine Learning Algorithm and Scientific Software Library /
Van Boxel, Danielle,

Bayesian Additive Regression Networks (BARN): A Machine Learning Algorithm and Scientific Software Library /Danielle Van Boxel. - 1 electronic resource (111 pages)

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

We apply Bayesian Additive Regression Tree (BART) principles to training an ensemble of small neural networks for regression tasks. Using Markov Chain Monte Carlo, we sample from the posterior distribution of neural networks that have a single hidden layer. To create an ensemble of these, we apply Gibbs' sampling to update each network against the residual target value (i.e. subtracting the effect of the other networks). We demonstrate the effectiveness of this technique on several benchmark regression problems, comparing it to equivalent shallow neural networks, BART, and ordinary least squares. Our Bayesian Additive Regression Networks (BARN) provide more consistent and often more accurate results. On test data benchmarks, BARN averaged between 5\\% to 20\\% lower root mean square error. This error performance does come at the cost, however, of greater computation time. BARN sometimes takes on the order of a minute where competing methods take a second or less. But, BARN without cross-validated hyperparameter tuning takes about the same amount of computation time as tuned other methods. Yet BARN is still typically more accurate.Additionally, we make BARN available as a Python package, `barmpy`, with documentation at `https://dvbuntu.github.io/barmpy/` for general machine learning practitioners. Our object-oriented design is compatible with SciKit-Learn, allowing usage of their tools like cross-validation. To ease learning to use `barmpy`, we produce a companion tutorial that expands on reference information in the documentation. Any interested user can `pip install barmpy` from the official PyPi repository.Finally, we expand BARN beyond regression to classification tasks. Much like in regression, we evaluate it against competing classification methods on benchmark data sets, real and synthetic. Here, we find that BARN is relatively best in specific scenarios, and even then by only a few percentage points as measured by test accuracy. This is, however, consistent with analysis of BART results in earlier research, where other methods were just as, if not more, accurate.BARN is a new method in machine learning with promising evidence for wide applicability. In addition to demonstrating its effectiveness in different situations, we outline how to apply similar principles to model types other than shallow neural networks. With rigorous statistical support and friendly software, BARN should be considered for new general purpose modeling problems.

English

ISBN: 9798382322223Subjects--Topical Terms:

377601
Applied mathematics.
Subjects--Index Terms:

Bayesian modeling

Bayesian Additive Regression Networks (BARN): A Machine Learning Algorithm and Scientific Software Library /
LDR:04009nmm a22004093i 4500 001 676311
005 20250420214340.5
006 m o d
007 cr|nu||||||||
008 250901s2024 miu||||||m |||||||eng d
020 $a 9798382322223
035 $a (MiAaPQD)AAI31241398
035 $a AAI31241398
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Van Boxel, Danielle, $e author. $0 (orcid)0009-0002-4091-9858 $3 990076
245 1 0 $a Bayesian Additive Regression Networks (BARN): A Machine Learning Algorithm and Scientific Software Library / $c Danielle Van Boxel.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (111 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
500 $a Advisors: Roman-Palacios, Cristian; Tang, Xueying Committee members: Maccabe, Arthur Barney; Zhang, Hao Helen; Kennedy, Tom.
502 $b Ph.D. $c The University of Arizona $d 2024.
520 $a We apply Bayesian Additive Regression Tree (BART) principles to training an ensemble of small neural networks for regression tasks. Using Markov Chain Monte Carlo, we sample from the posterior distribution of neural networks that have a single hidden layer. To create an ensemble of these, we apply Gibbs' sampling to update each network against the residual target value (i.e. subtracting the effect of the other networks). We demonstrate the effectiveness of this technique on several benchmark regression problems, comparing it to equivalent shallow neural networks, BART, and ordinary least squares. Our Bayesian Additive Regression Networks (BARN) provide more consistent and often more accurate results. On test data benchmarks, BARN averaged between 5\\% to 20\\% lower root mean square error. This error performance does come at the cost, however, of greater computation time. BARN sometimes takes on the order of a minute where competing methods take a second or less. But, BARN without cross-validated hyperparameter tuning takes about the same amount of computation time as tuned other methods. Yet BARN is still typically more accurate.Additionally, we make BARN available as a Python package, `barmpy`, with documentation at `https://dvbuntu.github.io/barmpy/` for general machine learning practitioners. Our object-oriented design is compatible with SciKit-Learn, allowing usage of their tools like cross-validation. To ease learning to use `barmpy`, we produce a companion tutorial that expands on reference information in the documentation. Any interested user can `pip install barmpy` from the official PyPi repository.Finally, we expand BARN beyond regression to classification tasks. Much like in regression, we evaluate it against competing classification methods on benchmark data sets, real and synthetic. Here, we find that BARN is relatively best in specific scenarios, and even then by only a few percentage points as measured by test accuracy. This is, however, consistent with analysis of BART results in earlier research, where other methods were just as, if not more, accurate.BARN is a new method in machine learning with promising evidence for wide applicability. In addition to demonstrating its effectiveness in different situations, we outline how to apply similar principles to model types other than shallow neural networks. With rigorous statistical support and friendly software, BARN should be considered for new general purpose modeling problems.
546 $a English
590 $a School code: 0009
650 4 $a Applied mathematics. $3 377601
650 4 $a Statistics. $3 182057
653 $a Bayesian modeling
653 $a Machine learning
653 $a Scientific software
690 $a 0364
690 $a 0463
710 2 $a The University of Arizona. $b Applied Mathematics. $e degree granting institution. $3 990077
720 1 $a Roman-Palacios, Cristian $e degree supervisor.
720 1 $a Tang, Xueying $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 85-11B.
790 $a 0009
791 $a Ph.D.
792 $a 2024
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31241398