Analiza dimenzionalno ograničenog genetskog programiranja uz prilagođene operatore

master's thesis

• Title (english): Analysis of dimensionally-aware genetic programming with adapted operators
• Author: Sandra Zurak
• 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: 2022-06-28, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Tema ovog rada je primjena dimenzionalno ograničenog genetskog programiranja na simboličku regresiju. Koriste se jednadžbe iz prirode kao skup podataka koji simbolička regresija prima. Poznata je informacija o mjernim jedinicama kako bi mjerna jedinica cilja bila u skladu s jedinicama ulaznih parametara. Korišteni operatori su operator zamjene stabla, operatori množenja i dijeljenja s cijelim brojem, operator zbrajanja i oduzimanja, operatori sinusa i kosinusa te operator korijenovanja. Algoritmima lokalne pretrage i evolucijskom strategijom dobiveni su brojevi evaluacija potrebni da se postigne najbolje rješenje. Analizirani su i uspoređeni dobiveni podaci.

Abstract (english)

The topic of this paper is the application of dimensionally limited genetic programming to symbolic regression. Equations from nature are used as a set of data that symbolic regression receives. Information on the units of measure is known so that the target unit of measure is consistent with the units of input parameters. The operators used are the tree replacement operator, the multiplication and division operators with integer, the addition and subtraction operator, the sine and cosine operators, and the rooting operator. Local search algorithms and evolutionary strategy have resulted in the numbers of evaluations needed to achieve the best solution. The obtained data were analyzed and compared.

• Keywords: genetsko programiranje; evolucijska strategija; AI-Feynman; lokalno pretraživanje; DAGP
• Keywords (english): genetic programming; evolutionary strategy; AI-Feynman; local search; DAGP
• Language: croatian
• URN:NBN: urn:nbn:hr:168:451389
• 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: 2023-01-14 22:05:17