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.

Proceedings of Machine Learning Research
Distributed and Automated Evolutionary Deep Architecture Learning with Unprecedented Scalability , Optimization for and with Machine Learning
40th International Conference on Machine Learning, PMLR 2023
Centrum Wiskunde & Informatica, Amsterdam (CWI), The Netherlands

Dushatskiy, A., Chebykin, A., Alderliesten, T., & Bosman, P. (2023). Multi-objective population based training. In Proceedings of the 40th International Conference on Machine Learning, PMLR (pp. 8969–8989).