Multi-Objective Population Based Training

arXiv (Cornell University)

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.

Authors 4

1. Arkadiy Dushatskiy

   Centrum Wiskunde & Informatica

   Affiliation as printed

   Centrum Wiskunde & Informatica , Amsterdam , the Netherlands

2. Alexander Chebykin

   Centrum Wiskunde & Informatica

   Affiliation as printed

   Centrum Wiskunde & Informatica , Amsterdam , the Netherlands

3. Tanja Alderliesten Aachen

   Leiden University Medical Center

   Affiliation as printed

   Leiden University Medical Center , Leiden , the Netherlands

4. Peter A. N. Bosman

   Delft University of Technology · Centrum Wiskunde & Informatica

   Affiliation as printed

   Centrum Wiskunde & Informatica , Amsterdam , the Netherlands

   Delft University of Technology , Delft , the Netherlands