| Title: | Using machine learning methods to assess module performance contribution in modular optimization frameworks |
|---|
| Authors: | ID Kostovska, Ana, Institut "Jožef Stefan" (Author)
ID Vermetten, Diederick (Author)
ID Korošec, Peter, Institut "Jožef Stefan" (Author)
ID Džeroski, Sašo, Institut "Jožef Stefan" (Author)
ID Doerr, Carola (Author)
ID Eftimov, Tome, Institut "Jožef Stefan" (Author) |
|---|
| Files: |  URL - Source URL, visit https://direct.mit.edu/evco/article-abstract/doi/10.1162/evco_a_00356/123899/Using-Machine-Learning-Methods-to-Assess-Module?redirectedFrom=fulltext
 PDF - Presentation file, download (1,37 MB)
MD5: 7D7A574A260449EFC752BD8129275844
|
|---|
| Language: | English |
|---|
| Typology: | 1.01 - Original Scientific Article |
|---|
| Organization: |  IJS - Jožef Stefan Institute
|
|---|
| Abstract: | Modular algorithm frameworks not only allow for combinations never tested in manually selected algorithm portfolios, but they also provide a structured approach to assess which algorithmic ideas are crucial for the observed performance of algorithms. In this study, we propose a methodology for analyzing the impact of the different modules on the overall performance. We consider modular frameworks for two widely used families of derivative-free black-box optimization algorithms, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) and differential evolution (DE). More specifically, we use performance data of 324 modCMA-ES and 576 modDE algorithm variants (with each variant corresponding to a specific configuration of modules) obtained on the 24 BBOB problems for 6 different runtime budgets in 2 dimensions. Our analysis of these data reveals that the impact of individual modules on overall algorithm performance varies significantly. Notably, among the examined modules, the elitism module in CMA-ES and the linear population size reduction module in DE exhibit the most significant impact on performance. Furthermore, our exploratory data analysis of problem landscape data suggests that the most relevant landscape features remain consistent regardless of the configuration of individual modules, but the influence that these features have on regression accuracy varies. In addition, we apply classifiers that exploit feature importance with respect to the trained models for performance prediction and performance data, to predict the modular configurations of CMA-ES and DE algorithm variants. The results show that the predicted configurations do not exhibit a statistically significant difference in performance compared to the true configurations, with the percentage varying depending on the setup (from 49.1% to 95.5% for modCMA and 21.7% to 77.1% for DE) |
|---|
| Keywords: | evolutionary computation, modular algorithm frameworks, DE |
|---|
| Publication status: | Published |
|---|
| Publication version: | Author Accepted Manuscript |
|---|
| Submitted for review: | 16.07.2023 |
|---|
| Article acceptance date: | 03.07.2024 |
|---|
| Publication date: | 17.10.2024 |
|---|
| Publisher: | Massachusetts Institute of Technology |
|---|
| Year of publishing: | 2024 |
|---|
| Number of pages: | 31 str. |
|---|
| Source: | ZDA |
|---|
| PID: | 20.500.12556/DiRROS-20968  |
|---|
| UDC: | 004.85 |
|---|
| ISSN on article: | 1530-9304 |
|---|
| DOI: | 10.1162/evco_a_00356 |
|---|
| COBISS.SI-ID: | 206993155 |
|---|
| Copyright: | © 2024 Massachusetts Institute of Technology |
|---|
| Note: | Nasl. z nasl. zaslona;
Opis vira z dne 10. 9. 2024;
|
|---|
| Publication date in DiRROS: | 11.12.2024 |
|---|
| Views: | 84 |
|---|
| Downloads: | 31 |
|---|
| Metadata: |    |
|---|
|
:
|
Copy citation |
|---|
| | | | Share: |  |
|---|
Hover the mouse pointer over a document title to show the abstract or click
on the title to get all document metadata. |
