Analiza mogućnosti graf neuronskih mreža za određivanje svojstava grafa

undergraduate thesis

• Title (english): Analyzing the capabilities of graph neural networks to determine graph properties
• Author: Lovro Raguž
• Mentor: Stjepan Bogdan (mentor)
• Committee member: Stjepan Bogdan (član povjerenstva)
• Committee member: Dean Martinović (član povjerenstva)
• Committee member: Antonella Barišić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-08-31, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Cilj ovog rada je istražiti sposobnosti graf neuronskih mreža (GNN) za određivanje svojstava grafa, primjerice minimalnog nezavisnog dominantnog skupa (MIDS). Proučava matematičku definiciju i svojstva MIDS-a, kao i postojeće metode za njegovo pronalaženje. Nadalje, upoznavamo se s osnovnim modelima graf neuronskih mreža i dostupnim bibliotekama s otvorenim izvornim kodom koje se mogu koristiti za njihovu implementaciju. Koristeći skup podataka koji se sastoji od velikog broja različitih grafova i njihovih MIDS, razvijamo programsku podršku za treniranje, podešavanje i validiranje odabranog modela GNN-a.3

Abstract (english)

The aim of this paper is to explore the capabilities of graph neural networks (GNNs) in determining graph properties, such as the minimum independent dominating set (MIDS). It examines the mathematical definition and properties of MIDS, as well as existing methods for finding it. Furthermore, we familiarize ourselves with basic graph neural network models and available open-source libraries that can be used for their implementation. Using a dataset consisting of a large number of diverse graphs and their MIDS, we develop software support for training, tuning, and validating the chosen GNN model.

• Keywords: minimalni nezavisni dominantni skup; graf neuronske mreže; pytorch
• Keywords (english): minimum independent dominating set; graph neural network; pytorch
• Language: croatian
• URN:NBN: urn:nbn:hr:168:756853
• 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: 2023-10-11 10:18:02