Multi-Objective Population Based Training

Arkadiy Dushatskiy, Alexander Chebykin, Tanja Alderliesten, Peter Bosman
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:8969-8989, 2023.

Abstract

Population Based Training (PBT) is an efficient hyperparameter optimization algorithm. PBT is a single-objective algorithm, but many real-world hyperparameter optimization problems involve two or more conflicting objectives. In this work, we therefore introduce a multi-objective version of PBT, MO-PBT. Our experiments on diverse multi-objective hyperparameter optimization problems (Precision/Recall, Accuracy/Fairness, Accuracy/Adversarial Robustness) show that MO-PBT outperforms random search, single-objective PBT, and the state-of-the-art multi-objective hyperparameter optimization algorithm MO-ASHA.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-dushatskiy23a, title = {Multi-Objective Population Based Training}, author = {Dushatskiy, Arkadiy and Chebykin, Alexander and Alderliesten, Tanja and Bosman, Peter}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {8969--8989}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/dushatskiy23a/dushatskiy23a.pdf}, url = {https://proceedings.mlr.press/v202/dushatskiy23a.html}, abstract = {Population Based Training (PBT) is an efficient hyperparameter optimization algorithm. PBT is a single-objective algorithm, but many real-world hyperparameter optimization problems involve two or more conflicting objectives. In this work, we therefore introduce a multi-objective version of PBT, MO-PBT. Our experiments on diverse multi-objective hyperparameter optimization problems (Precision/Recall, Accuracy/Fairness, Accuracy/Adversarial Robustness) show that MO-PBT outperforms random search, single-objective PBT, and the state-of-the-art multi-objective hyperparameter optimization algorithm MO-ASHA.} }
Endnote
%0 Conference Paper %T Multi-Objective Population Based Training %A Arkadiy Dushatskiy %A Alexander Chebykin %A Tanja Alderliesten %A Peter Bosman %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-dushatskiy23a %I PMLR %P 8969--8989 %U https://proceedings.mlr.press/v202/dushatskiy23a.html %V 202 %X Population Based Training (PBT) is an efficient hyperparameter optimization algorithm. PBT is a single-objective algorithm, but many real-world hyperparameter optimization problems involve two or more conflicting objectives. In this work, we therefore introduce a multi-objective version of PBT, MO-PBT. Our experiments on diverse multi-objective hyperparameter optimization problems (Precision/Recall, Accuracy/Fairness, Accuracy/Adversarial Robustness) show that MO-PBT outperforms random search, single-objective PBT, and the state-of-the-art multi-objective hyperparameter optimization algorithm MO-ASHA.
APA
Dushatskiy, A., Chebykin, A., Alderliesten, T. & Bosman, P.. (2023). Multi-Objective Population Based Training. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:8969-8989 Available from https://proceedings.mlr.press/v202/dushatskiy23a.html.

Related Material