File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Noise-Efficient Learning of Differentially Private Partitioning Machine Ensembles

TitleNoise-Efficient Learning of Differentially Private Partitioning Machine Ensembles
Authors
KeywordsDifferential privacy
Ensembles
Noise reduction
Issue Date2023
Citation
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2023, v. 13716 LNAI, p. 587-603 How to Cite?
AbstractDifferentially private decision tree algorithms have been popular since the introduction of differential privacy. While many private tree-based algorithms have been proposed for supervised learning tasks, such as classification, very few extend naturally to the semi-supervised setting. In this paper, we present a framework that takes advantage of unlabelled data to reduce the noise requirement in differentially private decision forests and improves their predictive performance. The main ingredients in our approach consist of a median splitting criterion that creates balanced leaves, a geometric privacy budget allocation technique, and a random sampling technique to compute the private splitting-point accurately. While similar ideas existed in isolation, their combination is new, and has several advantages: (1) The semi-supervised mode of operation comes for free. (2) Our framework is applicable in two different privacy settings: when label-privacy is required, and when privacy of the features is also required. (3) Empirical evidence on 18 UCI data sets and 3 synthetic data sets demonstrate that our algorithm achieves high utility performance compared to the current state of the art in both supervised and semi-supervised classification problems.
Persistent Identifierhttp://hdl.handle.net/10722/330024
ISSN
2023 SCImago Journal Rankings: 0.606
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorHuang, Zhanliang-
dc.contributor.authorLei, Yunwen-
dc.contributor.authorKabán, Ata-
dc.date.accessioned2023-08-09T03:37:15Z-
dc.date.available2023-08-09T03:37:15Z-
dc.date.issued2023-
dc.identifier.citationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2023, v. 13716 LNAI, p. 587-603-
dc.identifier.issn0302-9743-
dc.identifier.urihttp://hdl.handle.net/10722/330024-
dc.description.abstractDifferentially private decision tree algorithms have been popular since the introduction of differential privacy. While many private tree-based algorithms have been proposed for supervised learning tasks, such as classification, very few extend naturally to the semi-supervised setting. In this paper, we present a framework that takes advantage of unlabelled data to reduce the noise requirement in differentially private decision forests and improves their predictive performance. The main ingredients in our approach consist of a median splitting criterion that creates balanced leaves, a geometric privacy budget allocation technique, and a random sampling technique to compute the private splitting-point accurately. While similar ideas existed in isolation, their combination is new, and has several advantages: (1) The semi-supervised mode of operation comes for free. (2) Our framework is applicable in two different privacy settings: when label-privacy is required, and when privacy of the features is also required. (3) Empirical evidence on 18 UCI data sets and 3 synthetic data sets demonstrate that our algorithm achieves high utility performance compared to the current state of the art in both supervised and semi-supervised classification problems.-
dc.languageeng-
dc.relation.ispartofLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)-
dc.subjectDifferential privacy-
dc.subjectEnsembles-
dc.subjectNoise reduction-
dc.titleNoise-Efficient Learning of Differentially Private Partitioning Machine Ensembles-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-031-26412-2_36-
dc.identifier.scopuseid_2-s2.0-85151053253-
dc.identifier.volume13716 LNAI-
dc.identifier.spage587-
dc.identifier.epage603-
dc.identifier.eissn1611-3349-
dc.identifier.isiWOS:000999043700036-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats