kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
The Performance of Privacy Funnel Algorithm Based on the Information Bottleneck Method
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

This thesis delves into the utility of the Information Bottleneck method foraddressing the Privacy Funnel challenge, employing greedy algorithms as the locally optimalsolution to the non-convex optimization. The Information Bottleneck method, as a significantexpansion of rate-distortion theory in the realm of information theory, focuses on optimizingdata compression while reducing information distortion. The research targets a delicatebalance, required to minimize private data inference while ensuring the reservation of usefulinformation in the output. Applying log-loss as the optimization measure, a greedy algorithmstrategy is deployed to implement the optimal solution. The algorithm's effectiveness isdemonstrated through an empirical evaluation based on a selected dataset published by theUS Census Bureau.

Abstract [sv]

Denna avhandling utforskar användbarheten avinformationsflaskhalsmetoden för att ta itu med integritetsfiltreringsutmaningen, genomanvändning av giriga algoritmer. Denna metod, en betydande utvidgning av teorin om takt-distorsion inom informationsteorin, fokuserar på att optimera komprimering av uttrycksamtidigt som informationsförvrängning minimeras. Forskningen siktar på den känsligabalansen som krävs för att minimera slutsatser av privat data samtidigt som användbarinformation i utdata bevaras. Genom att använda log-förlust som optimeringskriterium,implementeras en strategi med giriga algoritmer för att hitta den optimala lösningen.Algoritmernas effektivitet visas genom en grafisk framställning av numeriska analyserbaserade på en verklig datamängd.

Place, publisher, year, edition, pages
2024. , p. 637-643
Series
TRITA-EECS-EX ; 2024:192
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:kth:diva-359427OAI: oai:DiVA.org:kth-359427DiVA, id: diva2:1933507
Supervisors
Examiners
Projects
Kandidatexamensarbete Elektroteknik EECS 2024Available from: 2025-01-31 Created: 2025-01-31

Open Access in DiVA

fulltext(121150 kB)14 downloads
File information
File name FULLTEXT01.pdfFile size 121150 kBChecksum SHA-512
82342f1408fe2aae929f55e76f2a176a8521cf94c0fe100464225724a9b74ddff6f61a0fb1cf957b5d6400be30877b9d0aec28080253ae7c1e09793536e10217
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 14 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 700 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf