Diversity with Cooperation: Ensemble Methods for Few-Shot Classification

Nikita Dvornik 1 Cordelia Schmid 1 Julien Mairal 1
1 Thoth - Apprentissage de modèles à partir de données massives
LJK - Laboratoire Jean Kuntzmann, Inria Grenoble - Rhône-Alpes
Abstract : Few-shot classification consists of learning a predictive model that is able to effectively adapt to a new class, given only a few annotated samples. To solve this challenging problem, meta-learning has become a popular paradigm that advocates the ability to "learn to adapt". Recent works have shown, however, that simple learning strategies without meta-learning could be competitive. In this paper, we go a step further and show that by addressing the fundamental high-variance issue of few-shot learning classifiers, it is possible to significantly outperform current meta-learning techniques. Our approach consists of designing an ensemble of deep networks to leverage the variance of the classifiers, and introducing new strategies to encourage the networks to cooperate, while encouraging prediction diversity. Evaluation is conducted on the mini-ImageNet and CUB datasets, where we show that even a single network obtained by distillation yields state-of-the-art results.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02080004
Contributor : Nikita Dvornik <>
Submitted on : Tuesday, March 26, 2019 - 1:46:23 PM
Last modification on : Monday, April 1, 2019 - 2:21:57 PM

Files

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02080004, version 1
  • ARXIV : 1903.11341

Collections

Citation

Nikita Dvornik, Cordelia Schmid, Julien Mairal. Diversity with Cooperation: Ensemble Methods for Few-Shot Classification. 2019. ⟨hal-02080004⟩

Share

Metrics

Record views

93

Files downloads

87