Meta Learning for Few-Shot One-class Classification

TL;DR

This paper introduces a meta-learning approach for few-shot one-class classification that learns feature representations from multiclass data, enabling effective classification with minimal target class examples.

Contribution

It presents a novel meta-learning framework that trains feature representations for one-class classification using multiclass data, with a simple Prototypical Networks variant performing comparably to more complex methods.

Findings

Achieved performance comparable to state-of-the-art traditional one-class classifiers.

Demonstrated the effectiveness of learned features over specific one-class algorithms.

Improved baseline results in few-shot one-class classification scenarios.

Abstract

We propose a method that can perform one-class classification given only a small number of examples from the target class and none from the others. We formulate the learning of meaningful features for one-class classification as a meta-learning problem in which the meta-training stage repeatedly simulates one-class classification, using the classification loss of the chosen algorithm to learn a feature representation. To learn these representations, we require only multiclass data from similar tasks. We show how the Support Vector Data Description method can be used with our method, and also propose a simpler variant based on Prototypical Networks that obtains comparable performance, indicating that learning feature representations directly from data may be more important than which one-class algorithm we choose. We validate our approach by adapting few-shot classification datasets to the few-shot one-class classification scenario, obtaining similar results to the state-of-the-art of traditional one-class classification, and that improves upon that of one-class classification baselines employed in the few-shot setting. Our code is available at https://github.com/gdahia/meta_occ