Subspace Regularizers for Few-Shot Class Incremental Learning

TL;DR

This paper introduces a simple subspace regularization method for few-shot class incremental learning, enabling logistic regression classifiers to outperform complex models by encouraging new class weights to stay close to existing class subspaces.

Contribution

The paper proposes a novel, simple subspace regularization scheme that improves few-shot incremental learning with logistic regression, outperforming state-of-the-art methods.

Findings

Outperforms state-of-the-art approaches by up to 22% on miniImageNet.

Incorporating class descriptions further improves accuracy by up to 2%.

Simple geometric regularization effectively enhances continual learning.

Abstract

Few-shot class incremental learning -- the problem of updating a trained classifier to discriminate among an expanded set of classes with limited labeled data -- is a key challenge for machine learning systems deployed in non-stationary environments. Existing approaches to the problem rely on complex model architectures and training procedures that are difficult to tune and re-use. In this paper, we present an extremely simple approach that enables the use of ordinary logistic regression classifiers for few-shot incremental learning. The key to this approach is a new family of subspace regularization schemes that encourage weight vectors for new classes to lie close to the subspace spanned by the weights of existing classes. When combined with pretrained convolutional feature extractors, logistic regression models trained with subspace regularization outperform specialized, state-of-the-art approaches to few-shot incremental image classification by up to 22% on the miniImageNet dataset. Because of its simplicity, subspace regularization can be straightforwardly extended to incorporate additional background information about the new classes (including class names and descriptions specified in natural language); these further improve accuracy by up to 2%. Our results show that simple geometric regularization of class representations offers an effective tool for continual learning.