ICML 2021spotlight115 citations

Learning a Universal Template for Few-shot Dataset Generalization

Eleni Triantafillou, Hugo Larochelle, Richard Zemel, Vincent Dumoulin

Abstract

Few-shot dataset generalization is a challenging variant of the well-studied few-shot classification problem where a diverse training set of several datasets is given, for the purpose of training an adaptable model that can then learn classes from \emph{new datasets} using only a few examples. To this end, we propose to utilize the diverse training set to construct a \emph{universal template}: a partial model that can define a wide array of dataset-specialized models, by plugging in appropriate components. For each new few-shot classification problem, our approach therefore only requires inferring a small number of parameters to insert into the universal template. We design a separate network that produces an initialization of those parameters for each given task, and we then fine-tune its proposed initialization via a few steps of gradient descent. Our approach is more parameter-efficient, scalable and adaptable compared to previous methods, and achieves the state-of-the-art on the challenging Meta-Dataset benchmark.

BibTeX
@InProceedings{pmlr-v139-triantafillou21a,
  title = 	 {Learning a Universal Template for Few-shot Dataset Generalization},
  author =       {Triantafillou, Eleni and Larochelle, Hugo and Zemel, Richard and Dumoulin, Vincent},
  booktitle = 	 {Proceedings of the 38th International Conference on Machine Learning},
  pages = 	 {10424--10433},
  year = 	 {2021},
  editor = 	 {Meila, Marina and Zhang, Tong},
  volume = 	 {139},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {18--24 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v139/triantafillou21a/triantafillou21a.pdf},
  url = 	 {https://proceedings.mlr.press/v139/triantafillou21a.html},
  abstract = 	 {Few-shot dataset generalization is a challenging variant of the well-studied few-shot classification problem where a diverse training set of several datasets is given, for the purpose of training an adaptable model that can then learn classes from \emph{new datasets} using only a few examples. To this end, we propose to utilize the diverse training set to construct a \emph{universal template}: a partial model that can define a wide array of dataset-specialized models, by plugging in appropriate components. For each new few-shot classification problem, our approach therefore only requires inferring a small number of parameters to insert into the universal template. We design a separate network that produces an initialization of those parameters for each given task, and we then fine-tune its proposed initialization via a few steps of gradient descent. Our approach is more parameter-efficient, scalable and adaptable compared to previous methods, and achieves the state-of-the-art on the challenging Meta-Dataset benchmark.}
}
Learning a Universal Template for Few-shot Dataset Generalization · ICML 2021