Bag of Baselines for Multi-Objective Joint Neural Architecture Search and Hyperparameter Optimization
Abstract
While both neural architecture search (NAS) and hyperparameter optimization (HPO) have been studied extensively in recent years, NAS methods typically assume fixed hyperparameters and vice versa. Furthermore, NAS has recently often been framed as a multi-objective optimization problem, in order to take, e.g., resource requirements into account. In this paper, we propose a set of methods that extend current approaches to jointly optimize neural architectures and hyperparameters with respect to multiple objectives. We hope that these methods will serve as simple baselines for future research on multi-objective joint NAS + HPO.
Cite
Text
Izquierdo et al. "Bag of Baselines for Multi-Objective Joint Neural Architecture Search and Hyperparameter Optimization." ICML 2021 Workshops: AutoML, 2021.Markdown
[Izquierdo et al. "Bag of Baselines for Multi-Objective Joint Neural Architecture Search and Hyperparameter Optimization." ICML 2021 Workshops: AutoML, 2021.](https://mlanthology.org/icmlw/2021/izquierdo2021icmlw-bag/)BibTeX
@inproceedings{izquierdo2021icmlw-bag,
title = {{Bag of Baselines for Multi-Objective Joint Neural Architecture Search and Hyperparameter Optimization}},
author = {Izquierdo, Sergio and Guerrero-Viu, Julia and Hauns, Sven and Miotto, Guilherme and Schrodi, Simon and Biedenkapp, André and Elsken, Thomas and Deng, Difan and Lindauer, Marius and Hutter, Frank},
booktitle = {ICML 2021 Workshops: AutoML},
year = {2021},
url = {https://mlanthology.org/icmlw/2021/izquierdo2021icmlw-bag/}
}