FastBO: Fast HPO and NAS with Adaptive Fidelity Identification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiang, Jiantong, Mian, Ajmal
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912010548019200
author Jiang, Jiantong
Mian, Ajmal
author_facet Jiang, Jiantong
Mian, Ajmal
contents Hyperparameter optimization (HPO) and neural architecture search (NAS) are powerful in attaining state-of-the-art machine learning models, with Bayesian optimization (BO) standing out as a mainstream method. Extending BO into the multi-fidelity setting has been an emerging research topic, but faces the challenge of determining an appropriate fidelity for each hyperparameter configuration to fit the surrogate model. To tackle the challenge, we propose a multi-fidelity BO method named FastBO, which adaptively decides the fidelity for each configuration and efficiently offers strong performance. The advantages are achieved based on the novel concepts of efficient point and saturation point for each configuration.We also show that our adaptive fidelity identification strategy provides a way to extend any single-fidelity method to the multi-fidelity setting, highlighting its generality and applicability.
format Preprint
id arxiv_https___arxiv_org_abs_2409_00584
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FastBO: Fast HPO and NAS with Adaptive Fidelity Identification
Jiang, Jiantong
Mian, Ajmal
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Hyperparameter optimization (HPO) and neural architecture search (NAS) are powerful in attaining state-of-the-art machine learning models, with Bayesian optimization (BO) standing out as a mainstream method. Extending BO into the multi-fidelity setting has been an emerging research topic, but faces the challenge of determining an appropriate fidelity for each hyperparameter configuration to fit the surrogate model. To tackle the challenge, we propose a multi-fidelity BO method named FastBO, which adaptively decides the fidelity for each configuration and efficiently offers strong performance. The advantages are achieved based on the novel concepts of efficient point and saturation point for each configuration.We also show that our adaptive fidelity identification strategy provides a way to extend any single-fidelity method to the multi-fidelity setting, highlighting its generality and applicability.
title FastBO: Fast HPO and NAS with Adaptive Fidelity Identification
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.00584