An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers
This article gives a concise overview of Bayesian sampling for neural networks, and then presents an extensive evaluation on a set of various benchmark classification problems. The main objective is to study the sensitivity of this scheme to changes in the prior distribution of the parameters and hy...
Gespeichert in:
Veröffentlicht in: | Neural networks 1999-06, Vol.12 (4), p.677-705 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 705 |
---|---|
container_issue | 4 |
container_start_page | 677 |
container_title | Neural networks |
container_volume | 12 |
creator | Husmeier, D. Penny, W.D. Roberts, S.J. |
description | This article gives a concise overview of Bayesian sampling for neural networks, and then presents an extensive evaluation on a set of various benchmark classification problems. The main objective is to study the sensitivity of this scheme to changes in the prior distribution of the parameters and hyperparameters, and to evaluate the efficiency of the so-called automatic relevance determination (ARD) method. The article concludes with a comparison of the achieved classification results with those obtained with (i) the evidence scheme and (ii) with non-Bayesian methods. |
doi_str_mv | 10.1016/S0893-6080(99)00020-9 |
format | Article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_27011575</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0893608099000209</els_id><sourcerecordid>1859399943</sourcerecordid><originalsourceid>FETCH-LOGICAL-c452t-7e4a3e36737e4f61c6e95446067d3de0bbe0ac7a85319297cc80c61122d07c533</originalsourceid><addsrcrecordid>eNqF0c1u1DAUBWALgehQeASQFwi1i4B_EjteoTKigFTEAlhbd5wbanDswU5azdvj6YzoDlb24rvX1jmEPOfsNWdcvfnKeiMbxXp2Zsw5Y0ywxjwgK95r0wjdi4dk9ZeckCel_KxI9a18TE64UEoorVckXESK09Zn7yBQvIGwwOxTpGmk72CHxUOkBaZt8PEHvfXzNb3ebbIf6OcUZ6RryCHRMWU6Z_BxjyIuue6KON-m_Iu6AKX40WMuT8mjEULBZ8fzlHy_fP9t_bG5-vLh0_riqnFtJ-ZGYwsSpdKy3kbFnULTta1iSg9yQLbZIAOnoe8kN8Jo53rmFOdCDEy7TspT8uqwd5vT7wXLbCdfHIYAEdNSrNCM8053FZ79E_K-M9IY0-53dgfqciol42i32U-Qd5Yzu2_E3jVi93FbY-xdI9bUuRfHJ5bNhMP91LGCCl4eAZTawZghOl_unWGt4W1lbw8Ma3A3NU1bnMfocPAZ3WyH5P_zkz_hJqgW</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1859399943</pqid></control><display><type>article</type><title>An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers</title><source>Elsevier ScienceDirect Journals Complete</source><creator>Husmeier, D. ; Penny, W.D. ; Roberts, S.J.</creator><creatorcontrib>Husmeier, D. ; Penny, W.D. ; Roberts, S.J.</creatorcontrib><description>This article gives a concise overview of Bayesian sampling for neural networks, and then presents an extensive evaluation on a set of various benchmark classification problems. The main objective is to study the sensitivity of this scheme to changes in the prior distribution of the parameters and hyperparameters, and to evaluate the efficiency of the so-called automatic relevance determination (ARD) method. The article concludes with a comparison of the achieved classification results with those obtained with (i) the evidence scheme and (ii) with non-Bayesian methods.</description><identifier>ISSN: 0893-6080</identifier><identifier>EISSN: 1879-2782</identifier><identifier>DOI: 10.1016/S0893-6080(99)00020-9</identifier><identifier>PMID: 12662677</identifier><language>eng</language><publisher>Oxford: Elsevier Ltd</publisher><subject>Applied sciences ; Automatic relevance determination ; Bayesian statistics ; Benchmarking ; Classification problems ; Electric, optical and optoelectronic circuits ; Electronics ; Evidence approximation ; Exact sciences and technology ; Gibbs sampling ; Hybrid Monte Carlo ; Information, signal and communications theory ; Neural networks ; Parameters and hyperparameters ; Prior and posterior distribution ; Signal processing ; Speech processing ; Telecommunications and information theory</subject><ispartof>Neural networks, 1999-06, Vol.12 (4), p.677-705</ispartof><rights>1999 Elsevier Science Ltd</rights><rights>1999 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c452t-7e4a3e36737e4f61c6e95446067d3de0bbe0ac7a85319297cc80c61122d07c533</citedby><cites>FETCH-LOGICAL-c452t-7e4a3e36737e4f61c6e95446067d3de0bbe0ac7a85319297cc80c61122d07c533</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/S0893-6080(99)00020-9$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,780,784,3550,27924,27925,45995</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=1904914$$DView record in Pascal Francis$$Hfree_for_read</backlink><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/12662677$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Husmeier, D.</creatorcontrib><creatorcontrib>Penny, W.D.</creatorcontrib><creatorcontrib>Roberts, S.J.</creatorcontrib><title>An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers</title><title>Neural networks</title><addtitle>Neural Netw</addtitle><description>This article gives a concise overview of Bayesian sampling for neural networks, and then presents an extensive evaluation on a set of various benchmark classification problems. The main objective is to study the sensitivity of this scheme to changes in the prior distribution of the parameters and hyperparameters, and to evaluate the efficiency of the so-called automatic relevance determination (ARD) method. The article concludes with a comparison of the achieved classification results with those obtained with (i) the evidence scheme and (ii) with non-Bayesian methods.</description><subject>Applied sciences</subject><subject>Automatic relevance determination</subject><subject>Bayesian statistics</subject><subject>Benchmarking</subject><subject>Classification problems</subject><subject>Electric, optical and optoelectronic circuits</subject><subject>Electronics</subject><subject>Evidence approximation</subject><subject>Exact sciences and technology</subject><subject>Gibbs sampling</subject><subject>Hybrid Monte Carlo</subject><subject>Information, signal and communications theory</subject><subject>Neural networks</subject><subject>Parameters and hyperparameters</subject><subject>Prior and posterior distribution</subject><subject>Signal processing</subject><subject>Speech processing</subject><subject>Telecommunications and information theory</subject><issn>0893-6080</issn><issn>1879-2782</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>1999</creationdate><recordtype>article</recordtype><recordid>eNqF0c1u1DAUBWALgehQeASQFwi1i4B_EjteoTKigFTEAlhbd5wbanDswU5azdvj6YzoDlb24rvX1jmEPOfsNWdcvfnKeiMbxXp2Zsw5Y0ywxjwgK95r0wjdi4dk9ZeckCel_KxI9a18TE64UEoorVckXESK09Zn7yBQvIGwwOxTpGmk72CHxUOkBaZt8PEHvfXzNb3ebbIf6OcUZ6RryCHRMWU6Z_BxjyIuue6KON-m_Iu6AKX40WMuT8mjEULBZ8fzlHy_fP9t_bG5-vLh0_riqnFtJ-ZGYwsSpdKy3kbFnULTta1iSg9yQLbZIAOnoe8kN8Jo53rmFOdCDEy7TspT8uqwd5vT7wXLbCdfHIYAEdNSrNCM8053FZ79E_K-M9IY0-53dgfqciol42i32U-Qd5Yzu2_E3jVi93FbY-xdI9bUuRfHJ5bNhMP91LGCCl4eAZTawZghOl_unWGt4W1lbw8Ma3A3NU1bnMfocPAZ3WyH5P_zkz_hJqgW</recordid><startdate>19990601</startdate><enddate>19990601</enddate><creator>Husmeier, D.</creator><creator>Penny, W.D.</creator><creator>Roberts, S.J.</creator><general>Elsevier Ltd</general><general>Elsevier Science</general><scope>IQODW</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>19990601</creationdate><title>An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers</title><author>Husmeier, D. ; Penny, W.D. ; Roberts, S.J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c452t-7e4a3e36737e4f61c6e95446067d3de0bbe0ac7a85319297cc80c61122d07c533</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>1999</creationdate><topic>Applied sciences</topic><topic>Automatic relevance determination</topic><topic>Bayesian statistics</topic><topic>Benchmarking</topic><topic>Classification problems</topic><topic>Electric, optical and optoelectronic circuits</topic><topic>Electronics</topic><topic>Evidence approximation</topic><topic>Exact sciences and technology</topic><topic>Gibbs sampling</topic><topic>Hybrid Monte Carlo</topic><topic>Information, signal and communications theory</topic><topic>Neural networks</topic><topic>Parameters and hyperparameters</topic><topic>Prior and posterior distribution</topic><topic>Signal processing</topic><topic>Speech processing</topic><topic>Telecommunications and information theory</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Husmeier, D.</creatorcontrib><creatorcontrib>Penny, W.D.</creatorcontrib><creatorcontrib>Roberts, S.J.</creatorcontrib><collection>Pascal-Francis</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>Neural networks</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Husmeier, D.</au><au>Penny, W.D.</au><au>Roberts, S.J.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers</atitle><jtitle>Neural networks</jtitle><addtitle>Neural Netw</addtitle><date>1999-06-01</date><risdate>1999</risdate><volume>12</volume><issue>4</issue><spage>677</spage><epage>705</epage><pages>677-705</pages><issn>0893-6080</issn><eissn>1879-2782</eissn><abstract>This article gives a concise overview of Bayesian sampling for neural networks, and then presents an extensive evaluation on a set of various benchmark classification problems. The main objective is to study the sensitivity of this scheme to changes in the prior distribution of the parameters and hyperparameters, and to evaluate the efficiency of the so-called automatic relevance determination (ARD) method. The article concludes with a comparison of the achieved classification results with those obtained with (i) the evidence scheme and (ii) with non-Bayesian methods.</abstract><cop>Oxford</cop><pub>Elsevier Ltd</pub><pmid>12662677</pmid><doi>10.1016/S0893-6080(99)00020-9</doi><tpages>29</tpages></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0893-6080 |
ispartof | Neural networks, 1999-06, Vol.12 (4), p.677-705 |
issn | 0893-6080 1879-2782 |
language | eng |
recordid | cdi_proquest_miscellaneous_27011575 |
source | Elsevier ScienceDirect Journals Complete |
subjects | Applied sciences Automatic relevance determination Bayesian statistics Benchmarking Classification problems Electric, optical and optoelectronic circuits Electronics Evidence approximation Exact sciences and technology Gibbs sampling Hybrid Monte Carlo Information, signal and communications theory Neural networks Parameters and hyperparameters Prior and posterior distribution Signal processing Speech processing Telecommunications and information theory |
title | An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T07%3A22%3A21IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=An%20empirical%20evaluation%20of%20Bayesian%20sampling%20with%20hybrid%20Monte%20Carlo%20for%20training%20neural%20network%20classifiers&rft.jtitle=Neural%20networks&rft.au=Husmeier,%20D.&rft.date=1999-06-01&rft.volume=12&rft.issue=4&rft.spage=677&rft.epage=705&rft.pages=677-705&rft.issn=0893-6080&rft.eissn=1879-2782&rft_id=info:doi/10.1016/S0893-6080(99)00020-9&rft_dat=%3Cproquest_cross%3E1859399943%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1859399943&rft_id=info:pmid/12662677&rft_els_id=S0893608099000209&rfr_iscdi=true |