Advances in MetaDL - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Advances in MetaDL

Adrian El Baz
  • Function : Author
Zhengying Liu
  • Function : Author
  • PersonId : 1038060
Sebastien Treguer
  • Function : Author
  • PersonId : 1092606

Abstract

To stimulate advances in metalearning using deep learning techniques (MetaDL), we organized in 2021 a challenge and an associated workshop. This paper presents the design of the challenge and its results, and summarizes presentations made at the workshop. The challenge focused on few-shot learning classification tasks of small images. Participants' code submissions were run in a uniform manner, under tight computational constraints. This put pressure on solution designs to use existing architecture backbones and/or pre-trained networks. Winning methods featured various classifiers trained on top of the second last layer of popular CNN backbones, fined-tuned on the meta-training data (not necessarily in an episodic manner), then trained on the labeled support and tested on the unlabeled query sets of the meta-test data.
Fichier principal
Vignette du fichier
Edito_paper_Meta_Learning_workshop_AAAI_2021.pdf (341.63 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03550011 , version 1 (31-01-2022)

Identifiers

Cite

Adrian El Baz, Isabelle Guyon, Zhengying Liu, Jan N van Rijn, Sebastien Treguer, et al.. Advances in MetaDL. AAAI 2021 challenge and workshop, Feb 2021, Virtual, France. ⟨hal-03550011⟩
51 View
64 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More