On the Fundamental Limits of Matrix Completion: Leveraging Hierarchical Similarity Graphs
We study a matrix completion problem which leverages a hierarchical structure of social similarity graphs as side information in the context of recommender systems. We assume that users are categorized into clusters, each of which comprises sub-clusters (or what we call "groups"). We consi...
Uloženo v:
| Vydáno v: | IEEE transactions on information theory Ročník 70; číslo 3; s. 2039 - 2075 |
|---|---|
| Hlavní autoři: | , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
New York
IEEE
01.03.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Témata: | |
| ISSN: | 0018-9448, 1557-9654 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | We study a matrix completion problem which leverages a hierarchical structure of social similarity graphs as side information in the context of recommender systems. We assume that users are categorized into clusters, each of which comprises sub-clusters (or what we call "groups"). We consider a hierarchical stochastic block model that well respects practically-relevant social graphs and follows a low-rank rating matrix model. Under this setting, we characterize the information-theoretic limit on the number of observed matrix entries (i.e., optimal sample complexity) as a function of the quality of graph side information (to be detailed) by proving sharp upper and lower bounds on the sample complexity. One important consequence of this result is that leveraging the hierarchical structure of similarity graphs yields a substantial gain in sample complexity relative to the one that simply identifies different groups without resorting to the relational structure across them. Another implication of the result is when the graph information is rich, the optimal sample complexity is proportional to the number of clusters, while it nearly stays constant as the number of groups in a cluster increases. We empirically demonstrate through extensive experiments that the proposed algorithm achieves the optimal sample complexity. |
|---|---|
| AbstractList | We study a matrix completion problem which leverages a hierarchical structure of social similarity graphs as side information in the context of recommender systems. We assume that users are categorized into clusters, each of which comprises sub-clusters (or what we call “groups”). We consider a hierarchical stochastic block model that well respects practically-relevant social graphs and follows a low-rank rating matrix model. Under this setting, we characterize the information-theoretic limit on the number of observed matrix entries (i.e., optimal sample complexity) as a function of the quality of graph side information (to be detailed) by proving sharp upper and lower bounds on the sample complexity. One important consequence of this result is that leveraging the hierarchical structure of similarity graphs yields a substantial gain in sample complexity relative to the one that simply identifies different groups without resorting to the relational structure across them. Another implication of the result is when the graph information is rich, the optimal sample complexity is proportional to the number of clusters, while it nearly stays constant as the number of groups in a cluster increases. We empirically demonstrate through extensive experiments that the proposed algorithm achieves the optimal sample complexity. |
| Author | Elmahdy, Adel Suh, Changho Ahn, Junhyung Mohajer, Soheil |
| Author_xml | – sequence: 1 givenname: Junhyung orcidid: 0009-0003-0345-1608 surname: Ahn fullname: Ahn, Junhyung email: tonyahn96@kaist.ac.kr organization: School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea – sequence: 2 givenname: Adel orcidid: 0000-0002-2977-2237 surname: Elmahdy fullname: Elmahdy, Adel email: adel@umn.edu organization: Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, USA – sequence: 3 givenname: Soheil orcidid: 0000-0003-2254-1652 surname: Mohajer fullname: Mohajer, Soheil email: soheil@umn.edu organization: Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, USA – sequence: 4 givenname: Changho orcidid: 0000-0002-3101-4291 surname: Suh fullname: Suh, Changho email: chsuh@kaist.ac.kr organization: School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea |
| BookMark | eNp9kDtPwzAUhS1UJEphZ2CwxJziR1wnbKiiDymoA2VgsuzEaV0lTrBdRP89rtoBMbDcl853r-65BgPbWQ3AHUZjjFH-uF6uxwQROqY0ZTkiF2CIGeNJPmHpAAwRwlmSp2l2Ba6938U2ZZgMwcfKwrDVcLa3lWy1DbKBhWlN8LCr4asMznzDadf2jQ6ms0-w0F_ayY2xG7gwsXLl1pQReotQI50JBzh3st_6G3BZy8br23MegffZy3q6SIrVfDl9LpKS5CQkTHGdVjVnFSs5woTLGDliVKu6IkrllapxlnGlKs7rOEuxzIlklKKSqIrQEXg47e1d97nXPohdt3c2nhTxAJ8gTngWVZOTqnSd907XojRBHl8KTppGYCSONopoozjaKM42RhD9AXtnWukO_yH3J8RorX_JKWcMY_oDEX6AMQ |
| CODEN | IETTAW |
| CitedBy_id | crossref_primary_10_1016_j_patcog_2024_110456 |
| Cites_doi | 10.1109/TNSE.2016.2634322 10.1145/1864708.1864736 10.1109/MC.2009.263 10.1007/s13278-013-0141-9 10.1214/15-aos1428 10.1137/100811404 10.1561/9781680834772 10.1109/TIT.2010.2054251 10.1109/TIT.2015.2490670 10.1145/3321386 10.1109/ICASSP.2010.5495899 10.1137/080738970 10.1109/CCNC.2006.1593032 10.1109/ALLERTON.2018.8636058 10.1109/JPROC.2009.2035722 10.1016/j.physa.2011.12.021 10.1145/1557019.1557067 10.1007/s10107-017-1181-0 10.1073/pnas.98.2.404 10.1609/aaai.v29i1.9153 10.5555/2969442.2969475 10.1145/1458082.1458205 10.1109/ACCESS.2019.2928130 10.1007/s10208-009-9045-5 10.1109/ISIT50566.2022.9834449 10.1109/TPDS.2012.192 10.1145/1571941.1571978 10.1109/ICDM.2017.74 10.1016/j.acha.2015.08.003 10.1109/TIT.2010.2046205 10.1109/ICASSP.2017.7952911 10.1109/TSP.2021.3052033 10.1109/TPAMI.2012.271 10.1109/FOCS.2015.47 10.1002/0471722154 10.1007/s12532-012-0044-1 10.1145/1639714.1639745 10.1145/1935826.1935877 10.1145/2365952.2365969 10.5555/2969442.2969624 10.1137/110845768 10.1109/TIT.2010.2044061 10.1146/annurev.soc.27.1.415 10.1109/TPAMI.2010.231 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 |
| DBID | 97E RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D |
| DOI | 10.1109/TIT.2023.3345902 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library CrossRef 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 |
| DatabaseTitle | CrossRef 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 |
| DatabaseTitleList | Technology Research Database |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Computer Science |
| EISSN | 1557-9654 |
| EndPage | 2075 |
| ExternalDocumentID | 10_1109_TIT_2023_3345902 10375511 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: KAIST grantid: N11230067 funderid: 10.13039/501100007107 – fundername: Korean Government [Ministry of Science and ICT (MSIT)], Development and Study of AI Technologies to Inexpensively Conform to Evolving Policy on Ethics grantid: 2022-0-00184 – fundername: Institute of Information and Communications Technology Planning and Evaluation (IITP) Grant – fundername: Korea Advanced Institute of Science and Technology (KAIST) Grand Challenge 30 Program funderid: 10.13039/501100007107 – fundername: National Science Foundation grantid: CCF-1749981 funderid: 10.13039/100000001 |
| GroupedDBID | -~X .DC 0R~ 29I 3EH 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACGOD ACIWK 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 F5P HZ~ H~9 IAAWW IBMZZ ICLAB IDIHD IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS RXW TAE TN5 VH1 VJK AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D |
| ID | FETCH-LOGICAL-c292t-5b7e4df75d5c70127a7017053ebfd2bb9dbf1887bbd77fbfd41a92a5330c2bd23 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 1 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001203295400012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0018-9448 |
| IngestDate | Sun Nov 30 05:01:23 EST 2025 Sat Nov 29 03:31:51 EST 2025 Tue Nov 18 22:05:35 EST 2025 Wed Aug 27 02:17:05 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 3 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c292t-5b7e4df75d5c70127a7017053ebfd2bb9dbf1887bbd77fbfd41a92a5330c2bd23 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| ORCID | 0009-0003-0345-1608 0000-0003-2254-1652 0000-0002-2977-2237 0000-0002-3101-4291 |
| PQID | 2927607278 |
| PQPubID | 36024 |
| PageCount | 37 |
| ParticipantIDs | crossref_primary_10_1109_TIT_2023_3345902 crossref_citationtrail_10_1109_TIT_2023_3345902 ieee_primary_10375511 proquest_journals_2927607278 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-03-01 |
| PublicationDateYYYYMMDD | 2024-03-01 |
| PublicationDate_xml | – month: 03 year: 2024 text: 2024-03-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on information theory |
| PublicationTitleAbbrev | TIT |
| PublicationYear | 2024 |
| 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 ref56 ref15 ref59 ref58 ref52 ref11 ref55 ref10 ref17 Ashtiani (ref53) ref16 Massa (ref14) ref19 ref18 Elmahdy (ref28); 33 ref51 ref46 ref45 ref48 ref47 ref42 ref41 ref44 Mohan (ref37) 2012; 11 ref43 ref49 ref7 ref9 ref4 ref3 Mazumdar (ref54) ref5 Wang (ref39) ref40 Kalofolias (ref8) 2014 ref35 ref36 ref31 ref30 ref33 ref32 ref2 ref1 ref38 Jo (ref27) 2020 van den Berg (ref21) 2017 Fazel (ref34) 2002 MacWilliams (ref57) 1977; 16 ref23 ref26 ref25 ref22 Li (ref6) ref29 Monti (ref20); 30 Ahn (ref24) Gao (ref50) 2017; 18 |
| References_xml | – year: 2014 ident: ref8 article-title: Matrix completion on graphs publication-title: arXiv:1408.1717 – ident: ref51 doi: 10.1109/TNSE.2016.2634322 – year: 2002 ident: ref34 article-title: Matrix rank minimization with applications – year: 2017 ident: ref21 article-title: Graph convolutional matrix completion publication-title: arXiv:1706.02263 – ident: ref5 doi: 10.1145/1864708.1864736 – ident: ref1 doi: 10.1109/MC.2009.263 – ident: ref3 doi: 10.1007/s13278-013-0141-9 – start-page: 1126 volume-title: Proc. Int. Joint Conf. Artif. Intell. (IJCAI) ident: ref6 article-title: Relation regularized matrix factorization – ident: ref58 doi: 10.1214/15-aos1428 – ident: ref36 doi: 10.1137/100811404 – year: 2020 ident: ref27 article-title: Discrete-valued preference estimation with graph side information publication-title: arXiv:2003.07040 – ident: ref55 doi: 10.1561/9781680834772 – ident: ref38 doi: 10.1109/TIT.2010.2054251 – ident: ref56 doi: 10.1109/TIT.2015.2490670 – ident: ref52 doi: 10.1145/3321386 – ident: ref42 doi: 10.1109/ICASSP.2010.5495899 – start-page: 121 volume-title: Proc. AAAI ident: ref14 article-title: Controversial users demand local trust metrics: An experimental study on Epinions.com community – ident: ref35 doi: 10.1137/080738970 – ident: ref15 doi: 10.1109/CCNC.2006.1593032 – ident: ref25 doi: 10.1109/ALLERTON.2018.8636058 – ident: ref46 doi: 10.1109/JPROC.2009.2035722 – ident: ref47 doi: 10.1016/j.physa.2011.12.021 – start-page: 4272 volume-title: Proc. Adv. Neural Inf. Process. Syst. (NeurIPS) ident: ref24 article-title: Binary rating estimation with graph side information – volume: 11 start-page: 3441 year: 2012 ident: ref37 article-title: Iterative reweighted algorithms for matrix rank minimization publication-title: J. Mach. Learn. Res. – ident: ref16 doi: 10.1145/1557019.1557067 – ident: ref45 doi: 10.1007/s10107-017-1181-0 – ident: ref48 doi: 10.1073/pnas.98.2.404 – ident: ref11 doi: 10.1609/aaai.v29i1.9153 – ident: ref23 doi: 10.5555/2969442.2969475 – ident: ref9 doi: 10.1145/1458082.1458205 – ident: ref30 doi: 10.1109/ACCESS.2019.2928130 – ident: ref31 doi: 10.1007/s10208-009-9045-5 – ident: ref29 doi: 10.1109/ISIT50566.2022.9834449 – ident: ref18 doi: 10.1109/TPDS.2012.192 – ident: ref10 doi: 10.1145/1571941.1571978 – start-page: 91 volume-title: Proc. Int. Conf. Mach. Learn. (ICML) ident: ref39 article-title: Rank-one matrix pursuit for matrix completion – ident: ref12 doi: 10.1109/ICDM.2017.74 – ident: ref40 doi: 10.1016/j.acha.2015.08.003 – start-page: 4682 volume-title: Proc. Adv. Neural Inf. Process. Syst. (NIPS) ident: ref54 article-title: Query complexity of clustering with side information – ident: ref32 doi: 10.1109/TIT.2010.2046205 – ident: ref13 doi: 10.1109/ICASSP.2017.7952911 – start-page: 3216 volume-title: Proc. Adv. Neural Inf. Process. Syst. (NIPS) ident: ref53 article-title: Clustering with same-cluster queries – volume: 30 start-page: 3697 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref20 article-title: Geometric matrix completion with recurrent multi-graph neural networks – ident: ref26 doi: 10.1109/TSP.2021.3052033 – ident: ref44 doi: 10.1109/TPAMI.2012.271 – ident: ref49 doi: 10.1109/FOCS.2015.47 – volume: 33 start-page: 9061 volume-title: Proc. Adv. Neural Inf. Process. Syst. (NeurIPS) ident: ref28 article-title: Matrix completion with hierarchical graph side information – ident: ref59 doi: 10.1002/0471722154 – ident: ref41 doi: 10.1007/s12532-012-0044-1 – ident: ref17 doi: 10.1145/1639714.1639745 – volume: 16 volume-title: The Theory Error Correcting Codes year: 1977 ident: ref57 – ident: ref7 doi: 10.1145/1935826.1935877 – ident: ref19 doi: 10.1145/2365952.2365969 – ident: ref22 doi: 10.5555/2969442.2969624 – ident: ref43 doi: 10.1137/110845768 – volume: 18 start-page: 1980 year: 2017 ident: ref50 article-title: Achieving optimal misclassification proportion in stochastic block models publication-title: J. Mach. Learn. Res. – ident: ref33 doi: 10.1109/TIT.2010.2044061 – ident: ref2 doi: 10.1146/annurev.soc.27.1.415 – ident: ref4 doi: 10.1109/TPAMI.2010.231 |
| SSID | ssj0014512 |
| Score | 2.4501178 |
| Snippet | We study a matrix completion problem which leverages a hierarchical structure of social similarity graphs as side information in the context of recommender... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 2039 |
| SubjectTerms | Algorithms Clustering algorithms Clusters Collaborative filtering Complexity Complexity theory Filtering graph side information Graphs Information theory Lower bounds matrix completion problem Recommender systems Similarity Sparse matrices |
| Title | On the Fundamental Limits of Matrix Completion: Leveraging Hierarchical Similarity Graphs |
| URI | https://ieeexplore.ieee.org/document/10375511 https://www.proquest.com/docview/2927607278 |
| Volume | 70 |
| WOSCitedRecordID | wos001203295400012&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: 1557-9654 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014512 issn: 0018-9448 databaseCode: RIE dateStart: 19630101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEF60eNCD1apYrbIHLx7SJptNtutNxFpBq2CFegrZR6BQEulD_PnObNJSEQVvIewsSWZndyYz832EXPBIRVEWGI8FVnjccN-T1pVYdQ2ih3SVUY5sQgwG3dFIPlfN6q4Xxlrris9sGy9dLt8UeoG_yjrY0wYnPAQ7m0LEZbPWKmXAo6CEBg_AgiHoWOYkfdkZ3g_bSBPeDkOOcCXfziBHqvJjJ3bHS6_-zwfbI7uVH0mvS8Xvkw2bN0h9ydFAK5NtkJ01wMED8vaUU_D4aA_7P0pYf-panGa0yOgjwvV_UpwEIbmL_Io-WFjqjsiI9sfYq-yoUyb0BYQgJgYXnt4h5PXskLz2boc3fa8iV_A0k2zuRUpYbjIRmUgLzD-nwkHrhFZlhikljcoC2IGUMkJkcI8HqWQpFqNqpgwLj0gtL3J7TGicghvnhzpWmc9tqFVqI2kMk6kGk49Zk3SWnzvRFfI4EmBMEheB-DIBBSWooKRSUJNcriTeS9SNP8YeokLWxpW6aJLWUqVJZZezBF5dxD74bN2TX8ROyTbMzssysxapzacLe0a29Md8PJueuyX3BeAg04E |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dSxtBEB9KWqg-NDUqTY3tPvTFh0vu9naz2b6JGCONsWCE-HTcfhwE5FJMIv75zuxdRCkKvh3Hzn3Nzu7MzczvB_BLSCNlkbiIJ15Fwok40j6UWA0coYcMjDOBbEJNJoPZTP-tm9VDL4z3PhSf-S4dhly-W9g1_SrrUU8b7vAY7HyUQvC4atd6ShoImVTg4AnaMIYdm6xkrHvT82mXiMK7aSoIsOTFLhRoVf5bi8MGM2y-89G-wpfak2THlep34IMvW9DcsDSw2mhbsP0McnAXbi5Lhj4fG1IHSAXsz0KT05ItCnZBgP0PjC5CoNyL8jcbe5zsgcqIjebUrRzIU27ZFQphVIxOPDsj0OvlHlwPT6cno6imV4gs13wVSaO8cIWSTlpFGehcBXCd1JvCcWO0M0WCa5AxTqkCz4kk1zynclTLjePpPjTKRem_Aevn6MjFqe2bIhY-tSb3UjvHdW7R6Pu8Db3N585sjT1OFBi3WYhBYp2hgjJSUFYrqA1HTxL_KtyNN8bukUKejat00YbORqVZbZnLDF9d9WP02gbfXxH7CZ9H04txNj6f_DmALbyTqIrOOtBY3a39IXyy96v58u5HmH6PCIfWyA |
| 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=On+the+Fundamental+Limits+of+Matrix+Completion%3A+Leveraging+Hierarchical+Similarity+Graphs&rft.jtitle=IEEE+transactions+on+information+theory&rft.au=Ahn%2C+Junhyung&rft.au=Elmahdy%2C+Adel&rft.au=Mohajer%2C+Soheil&rft.au=Suh%2C+Changho&rft.date=2024-03-01&rft.pub=The+Institute+of+Electrical+and+Electronics+Engineers%2C+Inc.+%28IEEE%29&rft.issn=0018-9448&rft.eissn=1557-9654&rft.volume=70&rft.issue=3&rft.spage=2039&rft_id=info:doi/10.1109%2FTIT.2023.3345902&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0018-9448&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0018-9448&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0018-9448&client=summon |