Scalable Node-Level Computation Kernels for Parallel Exact Inference

In this paper, we investigate data parallelism in exact inference with respect to arbitrary junction trees. Exact inference is a key problem in exploring probabilistic graphical models, where the computation complexity increases dramatically with clique width and the number of states of random varia...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on computers Jg. 59; H. 1; S. 103 - 115
Hauptverfasser: Yinglong Xia, Prasanna, V.K.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: IEEE 01.01.2010
Schlagworte:
ISSN:0018-9340
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract In this paper, we investigate data parallelism in exact inference with respect to arbitrary junction trees. Exact inference is a key problem in exploring probabilistic graphical models, where the computation complexity increases dramatically with clique width and the number of states of random variables. We study potential table representation and scalable algorithms for node-level primitives. Based on such node-level primitives, we propose computation kernels for evidence collection and evidence distribution. A data parallel algorithm for exact inference is presented using the proposed computation kernels. We analyze the scalability of node-level primitives, computation kernels, and the exact inference algorithm using the coarse-grained multicomputer (CGM) model. According to the analysis, we achieve O(Nd c w c Pi j=1 wc r C,j /P) local computation time and O(N) global communication rounds using P processors, 1 les P les max c PiPi j1 wc r C,j, where N is the number of cliques in the junction tree; d c is the clique degree; r C,j is the number of states of the jth random variable in C; wc is the clique width; and w s is the separator width. We implemented the proposed algorithm on state-of-the-art clusters. Experimental results show that the proposed algorithm exhibits almost linear scalability over a wide range.
AbstractList In this paper, we investigate data parallelism in exact inference with respect to arbitrary junction trees. Exact inference is a key problem in exploring probabilistic graphical models, where the computation complexity increases dramatically with clique width and the number of states of random variables. We study potential table representation and scalable algorithms for node-level primitives. Based on such node-level primitives, we propose computation kernels for evidence collection and evidence distribution. A data parallel algorithm for exact inference is presented using the proposed computation kernels. We analyze the scalability of node-level primitives, computation kernels, and the exact inference algorithm using the coarse-grained multicomputer (CGM) model. According to the analysis, we achieve O(Nd c w c Pi j=1 wc r C,j /P) local computation time and O(N) global communication rounds using P processors, 1 les P les max c PiPi j1 wc r C,j, where N is the number of cliques in the junction tree; d c is the clique degree; r C,j is the number of states of the jth random variable in C; wc is the clique width; and w s is the separator width. We implemented the proposed algorithm on state-of-the-art clusters. Experimental results show that the proposed algorithm exhibits almost linear scalability over a wide range.
Author Prasanna, V.K.
Yinglong Xia
Author_xml – sequence: 1
  surname: Yinglong Xia
  fullname: Yinglong Xia
  organization: Comput. Sci. Dept., Univ. of Southern California, Los Angeles, CA, USA
– sequence: 2
  givenname: V.K.
  surname: Prasanna
  fullname: Prasanna, V.K.
  organization: Ming Hsieh Dept. of Electr. Eng., Univ. of Southern California, Los Angeles, CA, USA
BookMark eNp1z01Lw0AQBuA9VLCtnjx6yV0SZzfZr6PEWotBBeM5bHYnENkmZRNF_70pEQ-Cp2HgeYd5V2TR9R0SckEhoRT0dZknDEAnFMSCLAGoinWawSlZDcMbAAgGekluX6zxpvYYPfYO4wI_0Ed5vz-8j2Zs-y56wNChH6KmD9GzCcb7CWw-jR2jXddgwM7iGTlpjB_w_Geuyevdpszv4-Jpu8tvitgyno6x5bW12lkuKOe1S7lQVCDjTGonU6gbyR1z1NTamYlQntUgFAMhFUgtXbomdL5rQz8MAZvKtvObYzCtryhUx-pVmVfH6tMupszVn8whtHsTvv7Rl7NuEfFXcqoyBSr9BqipZVM
CODEN ITCOB4
CitedBy_id crossref_primary_10_1109_TPDS_2024_3414177
crossref_primary_10_1109_TPDS_2011_278
crossref_primary_10_1007_s10766_013_0246_9
Cites_doi 10.1109/SUPERC.1994.344295
10.1109/ICPADS.2006.96
10.1016/B978-1-55860-332-5.50070-5
10.1111/j.2517-6161.1988.tb01721.x
10.1093/bioinformatics/17.suppl_1.S243
10.1016/j.parco.2007.11.004
10.1016/B978-0-444-88071-0.50023-0
10.1109/71.605771
10.1109/IPDPS.2008.4536192
10.1023/A:1009730122752
10.1109/5.92042
10.1109/SBAC-PAD.2007.18
10.1177/0037549707084939
10.1177/1094342005051196
10.1007/s11227-006-0005-4
10.1016/j.jcss.2007.02.007
10.1007/s11227-006-0009-0
10.1109/IPDPS.2007.370556
10.1016/S0888-613X(96)00069-2
10.1109/ICPADS.2006.97
10.1109/TC.2007.1068
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
DOI 10.1109/TC.2009.106
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE/IET Electronic Library (IEL) (UW System Shared)
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE/IET Electronic Library (IEL) (UW System Shared)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
EndPage 115
ExternalDocumentID 10_1109_TC_2009_106
5184808
Genre orig-research
GroupedDBID --Z
-DZ
-~X
.55
.DC
0R~
29I
3EH
3O-
4.4
5GY
5VS
6IK
85S
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABFSI
ABQJQ
ABVLG
ACGFO
ACIWK
ACNCT
AENEX
AETEA
AETIX
AGQYO
AGSQL
AHBIQ
AI.
AIBXA
AKJIK
AKQYR
ALLEH
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
E.L
EBS
EJD
HZ~
H~9
IAAWW
IBMZZ
ICLAB
IEDLZ
IFIPE
IFJZH
IPLJI
JAVBF
LAI
M43
MS~
MVM
O9-
OCL
P2P
PQQKQ
RIA
RIE
RNI
RNS
RXW
RZB
TAE
TN5
TWZ
UHB
UKR
UPT
VH1
X7M
XJT
XOL
XZL
YXB
YYQ
YZZ
ZCG
AAYXX
ABUFD
CITATION
ID FETCH-LOGICAL-c253t-c5bcc9dc56155bd356816e25279d730bf75d2d1ab9da615154b068206780797d3
IEDL.DBID RIE
ISICitedReferencesCount 9
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000272314600009&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0018-9340
IngestDate Sat Nov 29 08:10:06 EST 2025
Tue Nov 18 21:46:37 EST 2025
Wed Aug 27 02:49:02 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 1
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c253t-c5bcc9dc56155bd356816e25279d730bf75d2d1ab9da615154b068206780797d3
PageCount 13
ParticipantIDs ieee_primary_5184808
crossref_citationtrail_10_1109_TC_2009_106
crossref_primary_10_1109_TC_2009_106
PublicationCentury 2000
PublicationDate 2010-Jan.
2010-01-00
PublicationDateYYYYMMDD 2010-01-01
PublicationDate_xml – month: 01
  year: 2010
  text: 2010-Jan.
PublicationDecade 2010
PublicationTitle IEEE transactions on computers
PublicationTitleAbbrev TC
PublicationYear 2010
Publisher IEEE
Publisher_xml – name: IEEE
References ref13
ref12
ref15
ref14
ref31
ref30
ref11
Pennock (ref1)
ref10
ref32
ref2
Grama (ref17) 2003
Middleton (ref20) 1991; 30
Bjøndalen (ref16)
Xia (ref22) 2009
ref24
ref23
ref26
ref25
Russell (ref3) 2002
ref28
ref27
(ref18) 2009
Dean (ref29)
ref8
ref7
ref9
(ref21) 2008
ref4
(ref19) 2009
ref6
Yin (ref5)
References_xml – ident: ref8
  doi: 10.1109/SUPERC.1994.344295
– ident: ref10
  doi: 10.1109/ICPADS.2006.96
– ident: ref9
  doi: 10.1016/B978-1-55860-332-5.50070-5
– volume-title: technical report, Univ. of Southern California
  year: 2009
  ident: ref22
  article-title: Node Level Computation Kernels for Parallel Exact Inference
– ident: ref6
  doi: 10.1111/j.2517-6161.1988.tb01721.x
– volume-title: The Bayesian Network Toolbox for Matlab (BNT)
  year: 2008
  ident: ref21
– start-page: 431
  volume-title: Proc. 14th Ann. Conf. Uncertainty in Artificial Intelligence
  ident: ref1
  article-title: Logarithmic Time Parallel Bayesian Inference
– ident: ref4
  doi: 10.1093/bioinformatics/17.suppl_1.S243
– ident: ref13
  doi: 10.1016/j.parco.2007.11.004
– volume: 30
  start-page: 241
  year: 1991
  ident: ref20
  article-title: Probabilistic Diagnosis Using a Reformulation of the INTERNIST-1/QMR Knowledge Base
  publication-title: Medicine
– start-page: 1
  volume-title: Proc. Ninth Int’l Symp. Artificial Intelligence and Math
  ident: ref29
  article-title: Scalable Inference in Hierarchical Generative Models
– ident: ref12
  doi: 10.1016/B978-0-444-88071-0.50023-0
– ident: ref28
  doi: 10.1109/71.605771
– ident: ref24
  doi: 10.1109/IPDPS.2008.4536192
– start-page: 133
  volume-title: Proc. Conf. Communicating Process Architectures
  ident: ref16
  article-title: Configurable Collective Communication in LAM-MPI
– ident: ref2
  doi: 10.1023/A:1009730122752
– volume-title: USC Center for High-Performance Computing and Communications (HPCC)
  year: 2009
  ident: ref18
– volume-title: Intel’s Probabilistic Networks Library (PNL)
  year: 2009
  ident: ref19
– volume-title: Introduction to Parallel Computing
  year: 2003
  ident: ref17
– ident: ref11
  doi: 10.1109/5.92042
– ident: ref15
  doi: 10.1109/SBAC-PAD.2007.18
– ident: ref31
  doi: 10.1177/0037549707084939
– ident: ref14
  doi: 10.1177/1094342005051196
– ident: ref32
  doi: 10.1007/s11227-006-0005-4
– ident: ref25
  doi: 10.1016/j.jcss.2007.02.007
– ident: ref26
  doi: 10.1007/s11227-006-0009-0
– volume-title: Artificial Intelligence: A Modern Approach
  year: 2002
  ident: ref3
– start-page: 122
  volume-title: Proc. Poster Book of the Eighth Int’l Conf. Research in Computational Molecular Biology
  ident: ref5
  article-title: Parallel Data Mining of Bayesian Networks from Gene Expression Data
– ident: ref27
  doi: 10.1109/IPDPS.2007.370556
– ident: ref7
  doi: 10.1016/S0888-613X(96)00069-2
– ident: ref30
  doi: 10.1109/ICPADS.2006.97
– ident: ref23
  doi: 10.1109/TC.2007.1068
SSID ssj0006209
Score 1.981217
Snippet In this paper, we investigate data parallelism in exact inference with respect to arbitrary junction trees. Exact inference is a key problem in exploring...
SourceID crossref
ieee
SourceType Enrichment Source
Index Database
Publisher
StartPage 103
SubjectTerms Bayesian network
Computational modeling
Exact inference
Inference algorithms
junction tree
Junctions
message passing
node-level primitives
Particle separators
Program processors
Random variables
Title Scalable Node-Level Computation Kernels for Parallel Exact Inference
URI https://ieeexplore.ieee.org/document/5184808
Volume 59
WOSCitedRecordID wos000272314600009&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/IET Electronic Library (IEL) (UW System Shared)
  issn: 0018-9340
  databaseCode: RIE
  dateStart: 19680101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://ieeexplore.ieee.org/
  omitProxy: false
  ssIdentifier: ssj0006209
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEB5q8aAHq61ifbGHnsTVvDabPUptUSylYJXeyr4CQkklpuLPd3fzoIdevCVhAmEmszO7M998AAMT4aLYVwLzkDMcxTzFPBEMhwFTklORElcu-JjQ6TRZLNisBXcNFkZr7ZrP9L29dLV8tZYbe1T2QMx2JLHI3j1K4xKr1ay6cd3O4RsHDiOvwuL5HnuYD8u5lL6lNdqKPlt0Ki6ajDv_-45jOKqyRvRYmvkEWjrrQqdmZECVg3bhcGu8YA-e3owBLDQKTddK44ntD0LlS84e6FXnmYmNyCSuaMZzS6uyQqNfLgv0UgMBT-F9PJoPn3HFmoBlQMICSyKkNIomtuIoVGgnjMU6IAFlyrizSClRgfK5YIq7dCYSXuymuCceZVSFZ9DO1pk-B8T9gAoRUZW68ptiJnlKSSAYj8wyENE-3NbaXMpqpLhltlgt3dbCY8v50PJcMnMf92HQCH-VkzR2i_WswhuRStcXux9fwkFZzrdnIlfQLvKNvoZ9-VN8fuc37g_5A5kRuB8
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEB5KFdSD1VaxPvfQk7ia93aPUltaGkvBKr2VfQWEkkpsxZ_v7uZBD714S8IEwkxmZ3ZnvvkAOjrCBZErOWY-oziIWIJZl1Pse1QKRngS2nLBR0wmk-58Tqc1eKiwMEop23ymHs2lreXLldiYo7KnUG9HugbZu2eYswq0VrXuRmVDh6td2A-cAo3nOvRp1ssnU7qG2Ggr_mwRqth4Mmj870tO4LjIG9FzbuhTqKm0CY2SkwEVLtqEo60Bgy14edMmMOAoNFlJhWPTIYTyl6xF0FhlqY6OSKeuaMoyQ6yyRP1fJtZoVEIBz-B90J_1hrjgTcDCC_01FiEXQqs6NDVHLn0zYyxSXugRKrVD84SE0pMu41Qym9AE3InsHPeuQyiR_jnU01WqLgAx1yOcB0QmtgAnqU6fktDjlAV6IQhIG-5LbS5EMVTccFssF3Zz4dDFrGeYLqm-j9rQqYS_8lkau8VaRuGVSKHry92P7-BgOHuNF_FoMr6Cw7y4b05IrqG-zjbqBvbFz_rzO7u1f8sfYQS7aA
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=Scalable+Node-Level+Computation+Kernels+for+Parallel+Exact+Inference&rft.jtitle=IEEE+transactions+on+computers&rft.au=Xia%2C+Yinglong&rft.au=Prasanna%2C+Viktor+K.&rft.date=2010-01-01&rft.issn=0018-9340&rft.volume=59&rft.issue=1&rft.spage=103&rft.epage=115&rft_id=info:doi/10.1109%2FTC.2009.106&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TC_2009_106
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0018-9340&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0018-9340&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0018-9340&client=summon