On Cost-Efficient Learning of Data Dependency

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 139
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorJang, Hyeryungko
dc.contributor.authorSong, Hyungseokko
dc.contributor.authorYi, Yungko
dc.date.accessioned2022-06-26T01:01:28Z-
dc.date.available2022-06-26T01:01:28Z-
dc.date.created2022-02-06-
dc.date.created2022-02-06-
dc.date.issued2022-06-
dc.identifier.citationIEEE-ACM TRANSACTIONS ON NETWORKING, v.30, no.3, pp.1382 - 1394-
dc.identifier.issn1063-6692-
dc.identifier.urihttp://hdl.handle.net/10203/297074-
dc.description.abstractIn this paper, we consider the problem of learning a tree graph structure that represents the statistical data dependency among nodes for a set of data samples generated by nodes, which provides the basic structure to perform a probabilistic inference task. Inference in the data graph includes marginal inference and maximum a posteriori (MAP) estimation, and belief propagation (BP) is a commonly used algorithm to compute the marginal distribution of nodes via message-passing, incurring non-negligible amount of communication cost. We inevitably have the trade-off between the inference accuracy and the message-passing cost because the learned structure of data dependency and physical connectivity graph are often highly different. In this paper, we formalize this trade-off in an optimization problem which outputs the data dependency graph that jointly considers learning accuracy and message-passing costs. We focus on two popular implementations of BP, ASYNC-BP and SYNC-BP, which have different message-passing mechanisms and cost structures. In ASYNC-BP, we propose a polynomial-time learning algorithm that is optimal, motivated by finding a maximum weight spanning tree of a complete graph. In SYNC-BP, we prove the NP-hardness of the problem and propose a greedy heuristic. For both BP implementations, we quantify how the error probability that the learned cost-efficient data graph differs from the ideal one decays as the number of data samples grows, using the large deviation principle, which provides a guideline on how many samples are necessary to obtain a certain trade-off. We validate our theoretical findings through extensive simulations, which confirms that it has a good match.-
dc.languageEnglish-
dc.publisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC-
dc.titleOn Cost-Efficient Learning of Data Dependency-
dc.typeArticle-
dc.identifier.wosid000745465600001-
dc.identifier.scopusid2-s2.0-85123358934-
dc.type.rimsART-
dc.citation.volume30-
dc.citation.issue3-
dc.citation.beginningpage1382-
dc.citation.endingpage1394-
dc.citation.publicationnameIEEE-ACM TRANSACTIONS ON NETWORKING-
dc.identifier.doi10.1109/TNET.2022.3141128-
dc.contributor.localauthorYi, Yung-
dc.contributor.nonIdAuthorJang, Hyeryung-
dc.description.isOpenAccessN-
dc.type.journalArticleArticle-
dc.subject.keywordAuthorCosts-
dc.subject.keywordAuthorDistributed databases-
dc.subject.keywordAuthorInference algorithms-
dc.subject.keywordAuthorGraphical models-
dc.subject.keywordAuthorTask analysis-
dc.subject.keywordAuthorData models-
dc.subject.keywordAuthorTree graphs-
dc.subject.keywordAuthorGraph structure learning-
dc.subject.keywordAuthordistributed inference-
dc.subject.keywordAuthorsample complexity-
dc.subject.keywordAuthorlarge deviation principle-
dc.subject.keywordAuthorbelief propagation-
dc.subject.keywordPlusBELIEF-PROPAGATION-
dc.subject.keywordPlusDISTRIBUTED INFERENCE-
dc.subject.keywordPlusTRACKING-
dc.subject.keywordPlusPRODUCT-
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0