PS-AAS: Portfolio Selection for Automated Algorithm Selection in Black-Box Optimization

Ana Kostovska, Gjorgjina Cenikj, Diederick Vermetten, Anja Jankovic, Ana Nikolikj, Urban Skvorc, Peter Korosec, Carola Doerr, Tome Eftimov
Proceedings of the Second International Conference on Automated Machine Learning, PMLR 224:11/1-17, 2023.

Abstract

The performance of automated algorithm selection (AAS) strongly depends on the portfolio of algorithms to choose from. Selecting the portfolio is a non-trivial task that requires balancing the trade-off between the higher flexibility of large portfolios with the increased complexity of the AAS task. In practice, probably the most common way to choose the algorithms for the portfolio is a greedy selection of the algorithms that perform well in some reference tasks of interest. We set out in this work to investigate alternative, data-driven portfolio selection techniques. Our proposed method creates algorithm behavior meta-representations, constructs a graph from a set of algorithms based on their meta-representation similarity, and applies a graph algorithm to select a final portfolio of diverse, representative, and non-redundant algorithms. We evaluate two distinct meta-representation techniques (SHAP and performance2vec) for selecting complementary portfolios from a total of 324 different variants of CMA-ES for the task of optimizing the BBOB single-objective problems in dimensionalities 5 and 30 with different cut-off budgets. We test two types of portfolios: one related to overall algorithm behavior and the ‘personalized’ one (related to algorithm behavior per each problem separately). We observe that the approach built on the performance2vec-based representations favors small portfolios with negligible error in the AAS task relative to the virtual best solver from the selected portfolio, whereas the portfolios built from the SHAP-based representations gain from higher flexibility at the cost of decreased performance of the AAS. Across most considered scenarios, personalized portfolios yield comparable or slightly better performance than the classical greedy approach. They outperform the full portfolio in all scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v224-kostovska23a, title = {PS-AAS: Portfolio Selection for Automated Algorithm Selection in Black-Box Optimization}, author = {Kostovska, Ana and Cenikj, Gjorgjina and Vermetten, Diederick and Jankovic, Anja and Nikolikj, Ana and Skvorc, Urban and Korosec, Peter and Doerr, Carola and Eftimov, Tome}, booktitle = {Proceedings of the Second International Conference on Automated Machine Learning}, pages = {11/1--17}, year = {2023}, editor = {Faust, Aleksandra and Garnett, Roman and White, Colin and Hutter, Frank and Gardner, Jacob R.}, volume = {224}, series = {Proceedings of Machine Learning Research}, month = {12--15 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v224/kostovska23a/kostovska23a.pdf}, url = {https://proceedings.mlr.press/v224/kostovska23a.html}, abstract = {The performance of automated algorithm selection (AAS) strongly depends on the portfolio of algorithms to choose from. Selecting the portfolio is a non-trivial task that requires balancing the trade-off between the higher flexibility of large portfolios with the increased complexity of the AAS task. In practice, probably the most common way to choose the algorithms for the portfolio is a greedy selection of the algorithms that perform well in some reference tasks of interest. We set out in this work to investigate alternative, data-driven portfolio selection techniques. Our proposed method creates algorithm behavior meta-representations, constructs a graph from a set of algorithms based on their meta-representation similarity, and applies a graph algorithm to select a final portfolio of diverse, representative, and non-redundant algorithms. We evaluate two distinct meta-representation techniques (SHAP and performance2vec) for selecting complementary portfolios from a total of 324 different variants of CMA-ES for the task of optimizing the BBOB single-objective problems in dimensionalities 5 and 30 with different cut-off budgets. We test two types of portfolios: one related to overall algorithm behavior and the ‘personalized’ one (related to algorithm behavior per each problem separately). We observe that the approach built on the performance2vec-based representations favors small portfolios with negligible error in the AAS task relative to the virtual best solver from the selected portfolio, whereas the portfolios built from the SHAP-based representations gain from higher flexibility at the cost of decreased performance of the AAS. Across most considered scenarios, personalized portfolios yield comparable or slightly better performance than the classical greedy approach. They outperform the full portfolio in all scenarios.} }
Endnote
%0 Conference Paper %T PS-AAS: Portfolio Selection for Automated Algorithm Selection in Black-Box Optimization %A Ana Kostovska %A Gjorgjina Cenikj %A Diederick Vermetten %A Anja Jankovic %A Ana Nikolikj %A Urban Skvorc %A Peter Korosec %A Carola Doerr %A Tome Eftimov %B Proceedings of the Second International Conference on Automated Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Aleksandra Faust %E Roman Garnett %E Colin White %E Frank Hutter %E Jacob R. Gardner %F pmlr-v224-kostovska23a %I PMLR %P 11/1--17 %U https://proceedings.mlr.press/v224/kostovska23a.html %V 224 %X The performance of automated algorithm selection (AAS) strongly depends on the portfolio of algorithms to choose from. Selecting the portfolio is a non-trivial task that requires balancing the trade-off between the higher flexibility of large portfolios with the increased complexity of the AAS task. In practice, probably the most common way to choose the algorithms for the portfolio is a greedy selection of the algorithms that perform well in some reference tasks of interest. We set out in this work to investigate alternative, data-driven portfolio selection techniques. Our proposed method creates algorithm behavior meta-representations, constructs a graph from a set of algorithms based on their meta-representation similarity, and applies a graph algorithm to select a final portfolio of diverse, representative, and non-redundant algorithms. We evaluate two distinct meta-representation techniques (SHAP and performance2vec) for selecting complementary portfolios from a total of 324 different variants of CMA-ES for the task of optimizing the BBOB single-objective problems in dimensionalities 5 and 30 with different cut-off budgets. We test two types of portfolios: one related to overall algorithm behavior and the ‘personalized’ one (related to algorithm behavior per each problem separately). We observe that the approach built on the performance2vec-based representations favors small portfolios with negligible error in the AAS task relative to the virtual best solver from the selected portfolio, whereas the portfolios built from the SHAP-based representations gain from higher flexibility at the cost of decreased performance of the AAS. Across most considered scenarios, personalized portfolios yield comparable or slightly better performance than the classical greedy approach. They outperform the full portfolio in all scenarios.
APA
Kostovska, A., Cenikj, G., Vermetten, D., Jankovic, A., Nikolikj, A., Skvorc, U., Korosec, P., Doerr, C. & Eftimov, T.. (2023). PS-AAS: Portfolio Selection for Automated Algorithm Selection in Black-Box Optimization. Proceedings of the Second International Conference on Automated Machine Learning, in Proceedings of Machine Learning Research 224:11/1-17 Available from https://proceedings.mlr.press/v224/kostovska23a.html.

Related Material