Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
BMI: Bounded Mutual Information for Efficient Privacy-Preserving Feature Selection
RISE Research Institutes of Sweden, Digitala system, Industriella system.ORCID-id: 0000-0002-1954-760x
RISE Research Institutes of Sweden, Digitala system, Datavetenskap.ORCID-id: 0000-0001-6116-164X
RISE Research Institutes of Sweden, Digitala system, Datavetenskap.ORCID-id: 0000-0002-2772-4661
RISE Research Institutes of Sweden, Digitala system, Datavetenskap.
Vise andre og tillknytning
2024 (engelsk)Inngår i: Lecture Notes in Computer Science, ISSN 0302-9743, E-ISSN 1611-3349, Vol. 14983 LNCS, s. 353-373Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

We introduce low complexity bounds on mutual information for efficient privacy-preserving feature selection with secure multi-party computation (MPC). Considering a discrete feature with N possible values and a discrete label with M possible values, our approach requires O(N) multiplications as opposed to O(NM) in a direct MPC implementation of mutual information. Our experimental results show that for regression tasks, we achieve a computation speed up of over 1,000× compared to a straightforward MPC implementation of mutual information, while achieving similar accuracy for the downstream machine learning model.

sted, utgiver, år, opplag, sider
Springer Science and Business Media Deutschland GmbH , 2024. Vol. 14983 LNCS, s. 353-373
Emneord [en]
Differential privacy; Complexity bounds; Computation speed; Features selection; Lower complexity; Multiparty computation; Mutual informations; Privacy; Privacy preserving; Secure multi-party computation; Speed up
HSV kategori
Identifikatorer
URN: urn:nbn:se:ri:diva-76193DOI: 10.1007/978-3-031-70890-9_18Scopus ID: 2-s2.0-85204610017OAI: oai:DiVA.org:ri-76193DiVA, id: diva2:1914214
Konferanse
29th European Symposium on Research in Computer Security, ESORICS 2024. Bydgoszcz. 16 September 2024 through 20 September 2024
Merknad

This research is funded by the EU Horizon Europe project HARPOCRATES (Grant ID. 101069535) and H2020 project CONCORDIA (Grant ID. 830927). We thank Tuomas Karhu for preparing the SpO2 data as well as help and advice in the process. We would also like to thank the anonymous reviewers for their comments and suggested improvements.

Tilgjengelig fra: 2024-11-18 Laget: 2024-11-18 Sist oppdatert: 2024-11-18bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Andre lenker

Forlagets fulltekstScopus

Person

Eklund, DavidIacovazzi, AlfonsoWang, HanRaza, Shahid

Søk i DiVA

Av forfatter/redaktør
Eklund, DavidIacovazzi, AlfonsoWang, HanRaza, Shahid
Av organisasjonen
I samme tidsskrift
Lecture Notes in Computer Science

Søk utenfor DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric

doi
urn-nbn
Totalt: 13 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
v. 2.45.0