Zajedničko strojno učenje uz očuvanje privatnosti

undergraduate thesis

• Title (english): Privacy-preserving collaborative machine learning
• Author: Silvije Škudar
• Mentor: Ante Đerek (mentor)
• Committee member: Ante Đerek (predsjednik povjerenstva)
• Committee member: Martin Tutek (član povjerenstva)
• Committee member: Darijan Marčetić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U radu je opisan i implementiran protokol za zajedničko strojno učenje uz očuvanje privatnosti. Prilikom njegovog opisa, definirani su i kriptografski primitivi koji se koriste u njegovoj implementaciji. Takod̄er su se formalno definirala i analizirala njegovih ciljeva i sigurnosnih svojstava u kontekstu pasivnog i aktivnog suparnika. Dalje, pokazuju se neke osnovne tehnike strojnog učenja i njegova funkcionalnost je testirana na konkretnom primjeru logističke regresije. Konačno, opisuju se neki poznati napadi koji mogu narušiti njegovu funkcionalnost ili privatnost korisničkih podataka.

Abstract (english)

This thesis describes a privacy-preserving collaborative machine learning protocol and provides its implementation. Alongside its description, it also defines the cryptographic primitives used in its implementation. Also, it presents a formal definition and analysis of its goals and security properties in the context of passive and active adversaries. Further, common machine learning techniques are shown and the functionality of the protocol is tested on a concrete implementation of logistic regression. Finally, the thesis describes some known attacks that can compromise its functionality or the privacy of users' data.

• Keywords: strojno učenje; zajedničko učenje; privatnost
• Keywords (english): machine learning; collaborative learning; privacy
• Language: croatian
• URN:NBN: urn:nbn:hr:168:941591
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 22:27:05