Impementation of simultaneous application of several surrogate models to evolutionary optimization
| dc.contributor.advisor | Holeňa, Martin | |
| dc.contributor.author | Juranko, Ján | |
| dc.contributor.referee | Gemrot Jakub | |
| dc.date.accepted | ||
| dc.date.accessioned | 2017-06-07T16:19:49Z | |
| dc.date.available | 2017-06-07T16:19:49Z | |
| dc.date.issued | 2017-05-09 | |
| dc.description.abstract | Táto práca sa zameriava na súčasné použitie viacerých Gaussovských procesov (GP) ako náhradných modelov metódy Covariance Matrix Adaptation Evolution Strategy (CMA-ES), ktorá je významným algoritmom v oblasti black-box optimalizácie. Práca obsahuje implementáciu v prostredí MATLAB v spojení s knižnicou Gaussian Processes for Machine Learning (GPML) a benchmarkovými testami z platformy COmparing Continuous Optimisers (COCO). Taktiež sa zaoberá výberom modelov, ktoré budíu natrénované, ako aj určením najlepšieho z implementovaných algoritmov pre výber náhradného modelu, ktorý určuje model použitý pre budúcu generáciu CMA-ES. Práca obsahuje aj výsledky experimentov, ktoré preukázali zlepšenie celkového výkonu v porovnaní s použitím len jediného náhradného modelu. | cs |
| dc.description.abstract | This thesis is focused on using multiple Gaussian processes (GP) simultaneously as surrogate models for the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) method, which is an important algorithm in the black-box optimization. The thesis contains an implementation in the MATLAB enviroment, which uses the Gaussian Processes for Machine Learning (GPML) library and benchmark tests from the COmparing Continuous Optimisers (COCO) platform. It also investigates the process of choosing the right models, which will be trained, as well as algorithms for the selection of the best surrogate model, which will be used for the next generation of CMA-ES. The thesis also contains the results of performed experiments that prove the improvement of overall performance when compared to using only one surrogate model. | en |
| dc.identifier | KOS-695600135105 | |
| dc.identifier.uri | http://hdl.handle.net/10467/70405 | |
| dc.language.iso | ENG | |
| dc.publisher | České vysoké učení technické v Praze | cs |
| dc.publisher | Czech Technical University in Prague | en |
| dc.rights | A university thesis is a work protected by the Copyright Act of the Czech Republic. Extracts, copies and transcripts of the thesis are allowed for personal use only and at one`s own expense. The use of thesis should be in compliance with the Copyright Act. | en |
| dc.rights | Vysokoškolská závěrečná práce je dílo chráněné autorským zákonem. Je možné pořizovat z něj na své náklady a pro svoji osobní potřebu výpisy, opisy a rozmnoženiny. Jeho využití musí být v souladu s autorským zákonem v platném znění. | cs |
| dc.subject | black-box optimalizácia, CMA-ES, náhradné modely, Gaussovské procesy | cs |
| dc.subject | black-box optimization, CMA-ES, surrogate models, Gaussian processes | en |
| dc.title | Implementace současného použití několika náhradních modelů pro evoluční optimalizaci | cs |
| dc.title | Impementation of simultaneous application of several surrogate models to evolutionary optimization | en |
| dc.type | master thesis | en |
| dspace.entity.type | Publication | |
| relation.isAdvisorOfPublication | 6b5e39cd-43fc-40f9-84ed-b50faaa4a629 | |
| relation.isAdvisorOfPublication.latestForDiscovery | 6b5e39cd-43fc-40f9-84ed-b50faaa4a629 | |
| relation.isAuthorOfPublication | 46ae971f-243b-414a-8994-4bf4db7595cc | |
| relation.isAuthorOfPublication.latestForDiscovery | 46ae971f-243b-414a-8994-4bf4db7595cc | |
| theses.degree.discipline | Webové a softwarové inženýrství | cs |
| theses.degree.grantor | katedra softwarového inženýrství | cs |
| theses.degree.programme | Informatika | cs |
Files
Original bundle
1 - 3 of 3
- Name:
- F8-DP-2017-Juranko-Jan-thesis.pdf
- Size:
- 7.05 MB
- Format:
- Adobe Portable Document Format
- Description:
- PLNY_TEXT
- Name:
- F8-DP-2017-posudek-Holena_Martin.pdf
- Size:
- 111.76 KB
- Format:
- Adobe Portable Document Format
- Description:
- POSUDEK
- Name:
- F8-DP-2017-posudek-Gemrot_Jakub.pdf
- Size:
- 107.02 KB
- Format:
- Adobe Portable Document Format
- Description:
- POSUDEK