Latent Gaussian Model Boosting

Latent Gaussian models and boosting are widely used techniques in statistics and machine learning. Tree-boosting shows excellent prediction accuracy on many data sets, but potential drawbacks are that it assumes conditional independence of samples, produces discontinuous predictions for, e.g., spati...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transactions on pattern analysis and machine intelligence Ročník 45; číslo 2; s. 1894 - 1905
Hlavný autor: Sigrist, Fabio
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States IEEE 01.02.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Latent Gaussian models and boosting are widely used techniques in statistics and machine learning. Tree-boosting shows excellent prediction accuracy on many data sets, but potential drawbacks are that it assumes conditional independence of samples, produces discontinuous predictions for, e.g., spatial data, and it can have difficulty with high-cardinality categorical variables. Latent Gaussian models, such as Gaussian process and grouped random effects models, are flexible prior models which explicitly model dependence among samples and which allow for efficient learning of predictor functions and for making probabilistic predictions. However, existing latent Gaussian models usually assume either a zero or a linear prior mean function which can be an unrealistic assumption. This article introduces a novel approach that combines boosting and latent Gaussian models to remedy the above-mentioned drawbacks and to leverage the advantages of both techniques. We obtain increased prediction accuracy compared to existing approaches in both simulated and real-world data experiments.
AbstractList Latent Gaussian models and boosting are widely used techniques in statistics and machine learning. Tree-boosting shows excellent prediction accuracy on many data sets, but potential drawbacks are that it assumes conditional independence of samples, produces discontinuous predictions for, e.g., spatial data, and it can have difficulty with high-cardinality categorical variables. Latent Gaussian models, such as Gaussian process and grouped random effects models, are flexible prior models which explicitly model dependence among samples and which allow for efficient learning of predictor functions and for making probabilistic predictions. However, existing latent Gaussian models usually assume either a zero or a linear prior mean function which can be an unrealistic assumption. This article introduces a novel approach that combines boosting and latent Gaussian models to remedy the above-mentioned drawbacks and to leverage the advantages of both techniques. We obtain increased prediction accuracy compared to existing approaches in both simulated and real-world data experiments.
Latent Gaussian models and boosting are widely used techniques in statistics and machine learning. Tree-boosting shows excellent prediction accuracy on many data sets, but potential drawbacks are that it assumes conditional independence of samples, produces discontinuous predictions for, e.g., spatial data, and it can have difficulty with high-cardinality categorical variables. Latent Gaussian models, such as Gaussian process and grouped random effects models, are flexible prior models which explicitly model dependence among samples and which allow for efficient learning of predictor functions and for making probabilistic predictions. However, existing latent Gaussian models usually assume either a zero or a linear prior mean function which can be an unrealistic assumption. This article introduces a novel approach that combines boosting and latent Gaussian models to remedy the above-mentioned drawbacks and to leverage the advantages of both techniques. We obtain increased prediction accuracy compared to existing approaches in both simulated and real-world data experiments.Latent Gaussian models and boosting are widely used techniques in statistics and machine learning. Tree-boosting shows excellent prediction accuracy on many data sets, but potential drawbacks are that it assumes conditional independence of samples, produces discontinuous predictions for, e.g., spatial data, and it can have difficulty with high-cardinality categorical variables. Latent Gaussian models, such as Gaussian process and grouped random effects models, are flexible prior models which explicitly model dependence among samples and which allow for efficient learning of predictor functions and for making probabilistic predictions. However, existing latent Gaussian models usually assume either a zero or a linear prior mean function which can be an unrealistic assumption. This article introduces a novel approach that combines boosting and latent Gaussian models to remedy the above-mentioned drawbacks and to leverage the advantages of both techniques. We obtain increased prediction accuracy compared to existing approaches in both simulated and real-world data experiments.
Author Sigrist, Fabio
Author_xml – sequence: 1
  givenname: Fabio
  orcidid: 0000-0002-3994-2244
  surname: Sigrist
  fullname: Sigrist, Fabio
  email: fabio.sigrist@hslu.ch
  organization: Lucerne University of Applied Sciences and Arts, Rotkreuz, Switzerland
BackLink https://www.ncbi.nlm.nih.gov/pubmed/35439126$$D View this record in MEDLINE/PubMed
BookMark eNp9kD1PwzAQQC1URD_gD4BUVWJhSbHPsWOPpYJSqRUMZbYc54pSpXGJk4F_T0pLhw5MXt67O78-6ZS-REJuGR0zRvXj6n2ynI-BAow5k4oJuCA9YJJGGjR0SI8yCZFSoLqkH8KGUhYLyq9Il4uYawayR4YLW2NZj2a2CSG35WjpMyxGT96HOi8_r8nl2hYBb47vgHy8PK-mr9HibTafThaR41rUUaxEQiFhInUZ41KkDNVaaiodqgwxsykgxJkTHDjna8ZBSqmdSxAUd0nGB-ThMHdX-a8GQ222eXBYFLZE3wQDUoCSrZ606P0ZuvFNVbbXGUgk40msYtlSwyPVpFvMzK7Kt7b6Nn8_bwE4AK7yIVS4PiGMmn1e85vX7POaY95WUmeSy2tb576sK5sX_6t3BzVHxNMunQiteMx_AODghAY
CODEN ITPIDJ
CitedBy_id crossref_primary_10_1080_01621459_2024_2410004
crossref_primary_10_1016_j_ejor_2022_06_035
crossref_primary_10_1002_cncy_22725
crossref_primary_10_1016_j_aap_2025_108088
crossref_primary_10_1093_ibd_izaf060
crossref_primary_10_1111_ene_70124
crossref_primary_10_1016_j_ascom_2024_100818
crossref_primary_10_1016_j_ecolind_2025_114087
crossref_primary_10_1016_j_pirs_2024_100033
crossref_primary_10_1002_cncy_22732
crossref_primary_10_1017_asb_2024_7
crossref_primary_10_1016_j_trd_2025_104662
crossref_primary_10_1016_j_ufug_2025_128976
crossref_primary_10_1002_jvc2_219
crossref_primary_10_1038_s41598_023_28631_y
crossref_primary_10_1186_s12936_023_04760_7
crossref_primary_10_1002_ecs2_4751
crossref_primary_10_1016_j_ajpath_2023_02_020
crossref_primary_10_1080_13658816_2024_2348740
Cites_doi 10.1080/00949655.2012.741599
10.1080/03610918.2018.1490429
10.2307/143141
10.1073/pnas.1907378117
10.1109/TPAMI.2013.159
10.3758/s13428-017-0971-x
10.1016/j.eswa.2020.114080
10.3414/ME11-02-0021
10.1016/j.chemolab.2019.01.002
10.1198/016214503000125
10.1016/j.csda.2015.02.004
10.1007/s10994-019-05787-1
10.1515/ijb-2020-0136
10.17161/bi.v15i2.13384
10.1080/01621459.1986.10478240
10.2307/2699986
10.1214/11-STS361
10.1214/ss/1177013609
10.1214/009053606000000092
10.1080/01621459.2021.1950003
10.1007/s00180-012-0382-5
10.1198/016214503000170
10.1007/s10994-011-5258-3
10.2307/2337136
10.1073/pnas.1903070116
10.1007/978-1-4419-8853-9
10.1214/aos/1016218223
10.1002/sim.2738
10.1201/9781315370279
10.1016/j.spl.2017.02.033
10.1145/2939672.2939785
10.1145/3446776
10.1080/01621459.2020.1801451
10.1007/s11063-021-10434-9
10.1016/j.inffus.2021.11.011
10.1002/sam.11505
10.1109/CVPR.2011.5995605
10.1016/j.spl.2010.12.003
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
DBID 97E
ESBDL
RIA
RIE
AAYXX
CITATION
NPM
7SC
7SP
8FD
JQ2
L7M
L~C
L~D
7X8
DOI 10.1109/TPAMI.2022.3168152
DatabaseName IEEE Xplore (IEEE)
IEEE Xplore Open Access (Activated by CARLI)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
PubMed
Computer and Information Systems Abstracts
Electronics & Communications Abstracts
Technology Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
Technology Research Database
Computer and Information Systems Abstracts – Academic
Electronics & Communications Abstracts
ProQuest Computer Science Collection
Computer and Information Systems Abstracts
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitleList
PubMed
MEDLINE - Academic
Technology Research Database
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
EISSN 2160-9292
1939-3539
EndPage 1905
ExternalDocumentID 35439126
10_1109_TPAMI_2022_3168152
9759834
Genre orig-research
Journal Article
GrantInformation_xml – fundername: Swiss Innovation Agency
  grantid: 55463.1 IP-IC
GroupedDBID ---
-DZ
-~X
.DC
0R~
29I
4.4
53G
5GY
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACGFS
ACIWK
ACNCT
AENEX
AGQYO
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
E.L
EBS
EJD
ESBDL
F5P
HZ~
IEDLZ
IFIPE
IPLJI
JAVBF
LAI
M43
MS~
O9-
OCL
P2P
PQQKQ
RIA
RIE
RNS
RXW
TAE
TN5
UHB
~02
AAYXX
CITATION
5VS
9M8
AAYOK
ABFSI
ADRHT
AETEA
AETIX
AGSQL
AI.
AIBXA
ALLEH
FA8
H~9
IBMZZ
ICLAB
IFJZH
NPM
RIG
RNI
RZB
VH1
XJT
7SC
7SP
8FD
JQ2
L7M
L~C
L~D
7X8
ID FETCH-LOGICAL-c395t-485702715bcd1365b1e8f6906ce8deedab2e24dc532333f1326669cc7e283c7d3
IEDL.DBID RIE
ISICitedReferencesCount 19
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000912386000036&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0162-8828
1939-3539
IngestDate Thu Oct 02 11:51:29 EDT 2025
Sun Nov 30 05:19:22 EST 2025
Thu Apr 03 07:03:20 EDT 2025
Sat Nov 29 02:58:19 EST 2025
Tue Nov 18 22:24:37 EST 2025
Wed Aug 27 02:54:13 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 2
Language English
License https://creativecommons.org/licenses/by/4.0/legalcode
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c395t-485702715bcd1365b1e8f6906ce8deedab2e24dc532333f1326669cc7e283c7d3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ORCID 0000-0002-3994-2244
OpenAccessLink https://ieeexplore.ieee.org/document/9759834
PMID 35439126
PQID 2761374846
PQPubID 85458
PageCount 12
ParticipantIDs crossref_primary_10_1109_TPAMI_2022_3168152
pubmed_primary_35439126
proquest_miscellaneous_2652865327
crossref_citationtrail_10_1109_TPAMI_2022_3168152
proquest_journals_2761374846
ieee_primary_9759834
PublicationCentury 2000
PublicationDate 2023-02-01
PublicationDateYYYYMMDD 2023-02-01
PublicationDate_xml – month: 02
  year: 2023
  text: 2023-02-01
  day: 01
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
– name: New York
PublicationTitle IEEE transactions on pattern analysis and machine intelligence
PublicationTitleAbbrev TPAMI
PublicationTitleAlternate IEEE Trans Pattern Anal Mach Intell
PublicationYear 2023
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References ref13
ref12
ref15
ref52
ref11
ref10
ref17
ref19
ref18
Hothorn (ref23) 2010; 11
ref51
ref50
ref46
ref45
ref41
ref44
Wyner (ref42) 2017; 18
ref49
Pinheiro (ref7) 2006
Belkin (ref43)
ref9
ref4
Minka (ref35)
ref3
ref5
Williams (ref6) 2006
ref34
ref37
ref31
ref30
Freund (ref1)
ref33
Breiman (ref8) 1984
ref32
ref2
Lu (ref39) 2019
ref38
Kuss (ref36) 2005; 6
Rabinowicz (ref16) 2021
Ke (ref48)
ref24
ref26
ref25
Sigrist (ref14) 2020
ref20
Nesterov (ref40) 2004; 87
ref22
ref21
ref28
ref27
ref29
Guennebaud (ref47) 2010
References_xml – ident: ref13
  doi: 10.1080/00949655.2012.741599
– ident: ref20
  doi: 10.1080/03610918.2018.1490429
– ident: ref9
  doi: 10.2307/143141
– ident: ref45
  doi: 10.1073/pnas.1907378117
– start-page: 148
  volume-title: Proc. 13th Int. Conf. Mach. Learn.
  ident: ref1
  article-title: Experiments with a new boosting algorithm
– year: 2020
  ident: ref14
  article-title: Gaussian process boosting
– ident: ref5
  doi: 10.1109/TPAMI.2013.159
– volume-title: Classification and Regression Trees
  year: 1984
  ident: ref8
– ident: ref19
  doi: 10.3758/s13428-017-0971-x
– start-page: 362
  volume-title: Proc. 17th Conf. Uncertainty Artif. Intell.
  ident: ref35
  article-title: Expectation propagation for approximate Bayesian inference
– ident: ref37
  doi: 10.1016/j.eswa.2020.114080
– ident: ref27
  doi: 10.3414/ME11-02-0021
– ident: ref21
  doi: 10.1016/j.chemolab.2019.01.002
– ident: ref29
  doi: 10.1198/016214503000125
– ident: ref12
  doi: 10.1016/j.csda.2015.02.004
– ident: ref38
  doi: 10.1007/s10994-019-05787-1
– ident: ref15
  doi: 10.1515/ijb-2020-0136
– ident: ref51
  doi: 10.17161/bi.v15i2.13384
– year: 2019
  ident: ref39
  article-title: Accelerating gradient boosting machine
– year: 2021
  ident: ref16
  article-title: Trees-based models for correlated data
– volume: 11
  start-page: 2109
  issue: Aug
  year: 2010
  ident: ref23
  article-title: Model-based boosting 2.0
  publication-title: J. Mach. Learn. Res.
– ident: ref34
  doi: 10.1080/01621459.1986.10478240
– ident: ref2
  doi: 10.2307/2699986
– ident: ref33
  doi: 10.1214/11-STS361
– ident: ref24
  doi: 10.1214/ss/1177013609
– ident: ref28
  doi: 10.1214/009053606000000092
– ident: ref17
  doi: 10.1080/01621459.2021.1950003
– ident: ref50
  doi: 10.1007/s00180-012-0382-5
– ident: ref32
  doi: 10.1198/016214503000170
– ident: ref11
  doi: 10.1007/s10994-011-5258-3
– ident: ref46
  doi: 10.2307/2337136
– ident: ref44
  doi: 10.1073/pnas.1903070116
– volume: 87
  volume-title: Introductory Lectures on Convex Optimization: A Basic Course
  year: 2004
  ident: ref40
  doi: 10.1007/978-1-4419-8853-9
– start-page: 541
  volume-title: Proc. 35th Int. Conf. Mach. Learn.
  ident: ref43
  article-title: To understand deep learning we need to understand kernel learning
– ident: ref49
  doi: 10.1214/aos/1016218223
– ident: ref26
  doi: 10.1002/sim.2738
– volume-title: Mixed-Effects Models in S and S-PLUS
  year: 2006
  ident: ref7
– ident: ref25
  doi: 10.1201/9781315370279
– volume: 18
  start-page: 1
  issue: 48
  year: 2017
  ident: ref42
  article-title: Explaining the success of AdaBoost and random forests as interpolating classifiers
  publication-title: J. Mach. Learn. Res.
– ident: ref18
  doi: 10.1016/j.spl.2017.02.033
– year: 2010
  ident: ref47
  article-title: Eigen v3
– ident: ref3
  doi: 10.1145/2939672.2939785
– ident: ref41
  doi: 10.1145/3446776
– ident: ref52
  doi: 10.1080/01621459.2020.1801451
– volume-title: Gaussian Processes for Machine Learning
  year: 2006
  ident: ref6
– ident: ref31
  doi: 10.1007/s11063-021-10434-9
– volume: 6
  start-page: 1679
  issue: Oct
  year: 2005
  ident: ref36
  article-title: Assessing approximate inference for binary Gaussian process classification
  publication-title: J. Mach. Learn. Res.
– start-page: 3149
  volume-title: Proc. 31st Int. Conf. Neural Inf. Process. Syst.
  ident: ref48
  article-title: LightGBM: A highly efficient gradient boosting decision tree
– ident: ref4
  doi: 10.1016/j.inffus.2021.11.011
– ident: ref22
  doi: 10.1002/sam.11505
– ident: ref30
  doi: 10.1109/CVPR.2011.5995605
– ident: ref10
  doi: 10.1016/j.spl.2010.12.003
SSID ssj0014503
Score 2.5469508
Snippet Latent Gaussian models and boosting are widely used techniques in statistics and machine learning. Tree-boosting shows excellent prediction accuracy on many...
SourceID proquest
pubmed
crossref
ieee
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 1894
SubjectTerms Accuracy
Analytical models
Boosting
Data models
Gaussian process
gaussian processes
Machine learning
mixed effects models
Predictive models
Spatial data
Spatial databases
Splines (mathematics)
Title Latent Gaussian Model Boosting
URI https://ieeexplore.ieee.org/document/9759834
https://www.ncbi.nlm.nih.gov/pubmed/35439126
https://www.proquest.com/docview/2761374846
https://www.proquest.com/docview/2652865327
Volume 45
WOSCitedRecordID wos000912386000036&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 2160-9292
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014503
  issn: 0162-8828
  databaseCode: RIE
  dateStart: 19790101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEB5URPTgo76itUTwpmmbbLa7e6ziC7T0UKG3kGw2IJREbOvvd2bzoAcVvAWyeTCZYb7JznwfwBUiApVhmeAZX2ov9DPuKZ5i4ZpJbdJBqOPMUua_iNFITqdqvAY3zSyMMcY2n5kuHdq9_LTQS_pV1lOCK8nCdVgXQpSzWs2OQcitCjIiGIxwLCPqAZm-6k3Gw9dnLAWDoEsyTZixtmGLcZo5JU6FlXxkBVZ-x5o25zzs_e9t92G3wpbusHSGA1gzeQv2at0GtwrjFuyskBAeQucF4Wa-cB_j5ZwmKl2SR5u5t0Uxp5boI3h7uJ_cPXmVaoKnmeILLyTK-kD4PNEp9bAlvpEZ0RFrI1PMiHESmCBMNWcBYyzDahQrGKW1MIg0tEjZMWzkRW5OwU1Un4lskAqa1xV-Eie-L4xACCRkLPuJA35tu0hXlOKkbDGLbGnRV5E1fUSmjyrTO3DdXPNREmr8ufqQDNusrGzqQLv-RFEVc_MoEAhNiBp14MBlcxqjhbZA4twUS1wz4DSKywLhwEn5aZt71x5x9vMzz2GbpObLju02bCw-l-YCNvXX4n3-2UGXnMqOdclvv5jWvw
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1ZS8QwEB688HjwWo96VvBNq5uk2TSPKl64Lj6s4Ftp0xQEacXd9fc7kx74oIJvhSZtmWSYb5qZ7wM4RkSgc0wTAssiE4Qsl4GWGSaueWRs1gtNkjvK_L4aDKKXF_00BadtL4y11hWf2TO6dGf5WWkm9KvsXCupIxFOw6wMQ86qbq32zCCUTgcZMQz6OCYSTYtMV58Pny4e7zEZ5PyMhJowZi3CvJDUdUqsCt8ikpNY-R1tuqhzs_K_712F5Rpd-hfVdliDKVusw0qj3ODXjrwOS99oCDtw0EfAWYz922Qyop5KnwTS3vzLshxRUfQGPN9cD6_uglo3ITBCy3EQEmk9V0ymJqMqtpTZKCdCYmOjDGNiknLLw8xIwYUQOeajmMNoY5RFrGFUJjZhpigLuw1-qrtC5b1MUceuYmmSMqasQhCkoiTqph6wxnaxqUnFSdviLXbJRVfHzvQxmT6uTe_BSTvnvaLU-HN0hwzbjqxt6sFes0Rx7XWjmCsEJ0SO2vPgqL2N_kKHIElhywmO6UlqxhVcebBVLW377GZH7Pz8zkNYuBs-9uP-_eBhFxZJeL6q396DmfHHxO7DnPkcv44-DtzG_AIJL9ke
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Latent+Gaussian+Model+Boosting&rft.jtitle=IEEE+transactions+on+pattern+analysis+and+machine+intelligence&rft.au=Sigrist%2C+Fabio&rft.date=2023-02-01&rft.issn=1939-3539&rft.eissn=1939-3539&rft.volume=45&rft.issue=2&rft.spage=1894&rft_id=info:doi/10.1109%2FTPAMI.2022.3168152&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0162-8828&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0162-8828&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0162-8828&client=summon