Zaključivanje nad grafovima metodama dubokog učenja

master's thesis

• Title (english): Reasoning on Graphs based on Deep Learning Methods
• Author: Daniel Rakovac
• Mentor: Marko Čupić (mentor)
• Committee member: Marko Čupić (predsjednik povjerenstva)
• Committee member: Željka Mihajlović (član povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom je radu obrađena primjena dubokog učenja na zaključivanje nad grafovima. Na početku rada su objašnjeni osnovni pojmovi teorije grafova i ključni modeli dubokog učenja. Mnogi modeli definirani nad grafovima su samo modifikacije dubokih modela koje koristimo u ostalim područjima. Odabrana su tri modela za zaključivanje nad grafovima, to su konvolucijska mreža grafova, neuronska mreža grafova s pažnjom, te generativni suparnički model NetGAN. Primijenjeni su na problem klasifikacije čvorova i grafova, te na problem generiranja grafova. Za evaluaciju modela odabrane su tri citacijske mreže i skup grafova koji predstavljaju strukture proteina. Proveden je niz eksperimenata zbog ispitivanja utjecaja parametara na rezultate modela.

Abstract (english)

This paper describes the application of deep learning methods for reasoning on graphs. At the beginning of the paper, the base terms of graphs theory and key deep learning models are explained. Many models defined on graphs are just modifications of deep models used in other areas. Three models are selected for reasoning on graphs, convolutional graph network, graph attention network, and generative adversarial model NetGAN. They are applied to the problem of node and graph classification and graph generation. For evaluating the models, three citation networks and one graph set which contains protein structures are used. A series of experiments are conducted for exploring parameters influence on model results.

• Keywords: grafovi; duboko učenje; konvolucijska mreža grafova; neuronska mreža grafova s pažnjom; generativni suparnički modeli
• Keywords (english): graphs; deep learning; convolutional graph network; graph attention network; generative adversarial networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:199677
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 13:40:32