Bridging the Gap of AutoGraph Between Academia and Industry: Analyzing AutoGraph Challenge at KDD Cup 2020 - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2022

Bridging the Gap of AutoGraph Between Academia and Industry: Analyzing AutoGraph Challenge at KDD Cup 2020

Zhen Xu
  • Fonction : Auteur
  • PersonId : 1092603
Lanning Wei
  • Fonction : Auteur
  • PersonId : 1131637
Huan Zhao
  • Fonction : Auteur
  • PersonId : 1131638
Rex Ying
  • Fonction : Auteur
  • PersonId : 1131639
Wei-Wei Tu
  • Fonction : Auteur
  • PersonId : 1106352

Résumé

Graph structured data is ubiquitous in daily life and scientific areas and has attracted increasing attention. Graph Neural Networks (GNNs) have been proved to be effective in modeling graph structured data and many variants of GNN architectures have been proposed. However, much human effort is often needed to tune the architecture depending on different datasets. Researchers naturally adopt Automated Machine Learning on Graph Learning, aiming to reduce the human effort and achieve generally top-performing GNNs, but their methods focus more on the architecture search. To understand GNN practitioners' automated solutions, we organized AutoGraph Challenge at KDD Cup 2020, emphasizing on automated graph neural networks for node classification. We received top solutions especially from industrial tech companies like Meituan, Alibaba and Twitter, which are already open sourced on Github. After detailed comparisons with solutions from academia, we quantify the gaps between academia and industry on modeling scope, effectiveness and efficiency, and show that (1) academia AutoML for Graph solutions focus on GNN architecture search while industrial solutions, especially the winning ones in the KDD Cup, tend to obtain an overall solution (2) by neural architecture search only, academia solutions achieve on average 97.3% accuracy of industrial solutions (3) academia solutions are cheap to obtain with several GPU hours while industrial solutions take a few months' labors. Academic solutions also contain much fewer parameters.
Fichier principal
Vignette du fichier
AutoGraph_analysis.pdf (506.26 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03629994 , version 1 (04-04-2022)

Identifiants

  • HAL Id : hal-03629994 , version 1

Citer

Zhen Xu, Lanning Wei, Huan Zhao, Rex Ying, Quanming Yao, et al.. Bridging the Gap of AutoGraph Between Academia and Industry: Analyzing AutoGraph Challenge at KDD Cup 2020. 2022. ⟨hal-03629994⟩
92 Consultations
21 Téléchargements

Partager

Gmail Facebook X LinkedIn More