Primjena strojnog učenja u optimizaciji prstometa za klavir

master's thesis

• Title (english): Machine learning in piano fingering optimization
• Author: Dominik Milde
• Mentor: Domagoj Jakobović (mentor)
• Committee member: Domagoj Jakobović (predsjednik povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Committee member: Leonardo Jelenković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-07, Croatia

Abstract

U ovom radu problem optimizacije klavirskog prstometa opisan je kao problem strojnog učenja. Rad pruža uvid u dosadašnje pristupe rješavanja ovog problema. Opisani su osnovni koncepti strojnog učenja, neuronskih mreža na grafovima, sequence-to-sequence modela te korištenih skupovi podataka - ThumbSet i Piano Fingering Dataset. Ostvaren je programski okvir za ispitivanje postupaka strojnog učenja u primjeni na navedeni problem. Kroz eksperimente ocijenjena je učinkovitost postupaka strojnog učenja s obzirom na parametre algoritma učenja i raspoložive skupove podataka.

Abstract (english)

In this paper, the problem of piano fingering optimization is described as a machine learning problem. The work provides an insight into the current approaches to solving this problem. The basic concepts of machine learning, neural networks on graphs, sequence-to-sequence models and the data sets used - ThumbSet and Piano Fingering Dataset are described. A program framework for testing machine learning procedures in application to the stated problem was created. Through experiments, the effectiveness of machine learning procedures was evaluated with regard to the parameters of the learning algorithm and the available datasets.

• Keywords: strojno učenje; optimizacija klavirskog prstometa; Piano Fingering Dataset; ThumbSet; neuronske mreže na grafovima; sequence-to-sequence model; automatsko označavanje prstometa klavira
• Keywords (english): machine learning; piano fingering optimization; Piano Fingering Dataset; ThumbSet; graph neural networks; sequence-to-sequence model; automatic piano fingering
• Language: croatian
• URN:NBN: urn:nbn:hr:168:275293
• 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: 2023-10-11 09:50:35