Best Practices for Scientific Research on Neural Architecture Search

verfasst von

Marius Lindauer, Frank Hutter

Abstract

Finding a well-performing architecture is often tedious for both deep learning practitioners and researchers, leading to tremendous interest in the automation of this task by means of neural architecture search (NAS). Although the community has made major strides in developing better NAS methods, the quality of scientific empirical evaluations in the young field of NAS is still lacking behind that of other areas of machine learning. To address this issue, we describe a set of possible issues and ways to avoid them, leading to the NAS best practices checklist available at automl.org/nas_checklist.pdf.

Organisationseinheit(en)

Fachgebiet Maschinelles Lernen
Institut für Informationsverarbeitung

Externe Organisation(en)

Albert-Ludwigs-Universität Freiburg

Typ

Artikel

Journal

Journal of Machine Learning Research

Band

21

Anzahl der Seiten

18

ISSN

1532-4435

Publikationsdatum

11.2020

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Software, Artificial intelligence, Steuerungs- und Systemtechnik, Statistik und Wahrscheinlichkeit

Elektronische Version(en)

https://arxiv.org/abs/1909.02453 (Zugang: Offen)
https://jmlr.csail.mit.edu/papers/volume21/20-056/20-056.pdf (Zugang: Offen)