Self-Supervised Prototypical Transfer Learning for Few-Shot Classification

06/19/2020
by   Carlos Medina, et al.
21

Most approaches in few-shot learning rely on costly annotated data related to the goal task domain during (pre-)training. Recently, unsupervised meta-learning methods have exchanged the annotation requirement for a reduction in few-shot classification performance. Simultaneously, in settings with realistic domain shift, common transfer learning has been shown to outperform supervised meta-learning. Building on these insights and on advances in self-supervised learning, we propose a transfer learning approach which constructs a metric embedding that clusters unlabeled prototypical samples and their augmentations closely together. This pre-trained embedding is a starting point for few-shot classification by summarizing class clusters and fine-tuning. We demonstrate that our self-supervised prototypical transfer learning approach ProtoTransfer outperforms state-of-the-art unsupervised meta-learning methods on few-shot tasks from the mini-ImageNet dataset. In few-shot experiments with domain shift, our approach even has comparable performance to supervised methods, but requires orders of magnitude fewer labels.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/21/2018

A Meta-Learning Approach for Custom Model Training

Transfer-learning and meta-learning are two effective methods to apply k...
research
04/15/2022

Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference

Few-shot learning (FSL) is an important and topical problem in computer ...
research
04/28/2020

Meta-Learning for Few-Shot Land Cover Classification

The representations of the Earth's surface vary from one geographic regi...
research
01/03/2021

Few-shot Image Classification: Just Use a Library of Pre-trained Feature Extractors and a Simple Classifier

Recent papers have suggested that transfer learning can outperform sophi...
research
11/24/2019

Invenio: Discovering Hidden Relationships Between Tasks/Domains Using Structured Meta Learning

Exploiting known semantic relationships between fine-grained tasks is cr...
research
07/22/2020

CrossTransformers: spatially-aware few-shot transfer

Given new tasks with very little data–such as new classes in a classific...
research
02/22/2019

Centroid Networks for Few-Shot Clustering and Unsupervised Few-Shot Classification

Traditional clustering algorithms such as K-means rely heavily on the na...

Please sign up or login with your details

Forgot password? Click here to reset