%0 Conference Paper %1 06752577a51d4dc6b6f5baeaff8cba55 %A Benjamins, Carolin %A Raponi, Elena %A Jankovic, Anja %A Doerr, Carola %A Lindauer, Marius %B GECCO '23: Proceedings of the Genetic and Evolutionary Computation Conference Companion %D 2023 %I Association for Computing Machinery Special Interest Group on Genetic and Evolutionary Computation (SIGEVO) %K #sys:relevantfor:l3s automl myown %T Towards Self-Adjusting Weighted Expected Improvement for Bayesian Optimization %X In optimization, we often encounter expensive black-box problemswith unknown problem structures. Bayesian Optimization (BO) isa popular, surrogate-assisted and thus sample-efficient approachfor this setting. The BO pipeline itself is highly configurable withmany different design choices regarding the initial design, surrogatemodel and acquisition function (AF). Unfortunately, our understand-ing of how to select suitable components for a problem at hand isvery limited. In this work, we focus on the choice of the AF, whosemain purpose it is to balance the trade-off between exploring re-gions with high uncertainty and those with high promise for goodsolutions. We propose Self-Adjusting Weighted Expected Improve-ment (SAWEI), where we let the exploration-exploitation trade-offself-adjust in a data-driven manner based on a convergence crite-rion for BO. On the BBOB functions of the COCO benchmark, ourmethod performs favorably compared to handcrafted baselines andserves as a robust default choice for any problem structure. WithSAWEI, we are a step closer to on-the-fly, data-driven and robustBO designs that automatically adjust their sampling behavior tothe problem at hand.

@inproceedings{06752577a51d4dc6b6f5baeaff8cba55, abstract = {In optimization, we often encounter expensive black-box problemswith unknown problem structures. Bayesian Optimization (BO) isa popular, surrogate-assisted and thus sample-efficient approachfor this setting. The BO pipeline itself is highly configurable withmany different design choices regarding the initial design, surrogatemodel and acquisition function (AF). Unfortunately, our understand-ing of how to select suitable components for a problem at hand isvery limited. In this work, we focus on the choice of the AF, whosemain purpose it is to balance the trade-off between exploring re-gions with high uncertainty and those with high promise for goodsolutions. We propose Self-Adjusting Weighted Expected Improve-ment (SAWEI), where we let the exploration-exploitation trade-offself-adjust in a data-driven manner based on a convergence crite-rion for BO. On the BBOB functions of the COCO benchmark, ourmethod performs favorably compared to handcrafted baselines andserves as a robust default choice for any problem structure. WithSAWEI, we are a step closer to on-the-fly, data-driven and robustBO designs that automatically adjust their sampling behavior tothe problem at hand.}, added-at = {2024-02-14T12:39:33.000+0100}, author = {Benjamins, Carolin and Raponi, Elena and Jankovic, Anja and Doerr, Carola and Lindauer, Marius}, biburl = {https://www.bibsonomy.org/bibtex/2e2eed2dabee8c25069d2f61b1f6ba491/ail3s}, booktitle = {GECCO '23: Proceedings of the Genetic and Evolutionary Computation Conference Companion}, interhash = {6966fe4bffaf005faa6e173ee8f02c59}, intrahash = {e2eed2dabee8c25069d2f61b1f6ba491}, keywords = {#sys:relevantfor:l3s automl myown}, language = {English}, publisher = {Association for Computing Machinery Special Interest Group on Genetic and Evolutionary Computation (SIGEVO)}, timestamp = {2024-02-27T12:14:51.000+0100}, title = {Towards Self-Adjusting Weighted Expected Improvement for Bayesian Optimization}, year = 2023 }