Better Generalized Few-Shot Learning Even Without Base Data

TL;DR

This paper proposes a novel zero-base generalized few-shot learning method that effectively incorporates knowledge of novel classes without using any base data, outperforming existing methods that rely on base data.

Contribution

It introduces a normalization technique that controls both mean and variance of weight distributions for novel classes without base data, advancing GFSL research.

Findings

Zero-base GFSL outperforms existing GFSL methods using base data.

The proposed normalization improves weight distribution for novel classes.

Method is effective without access to base class samples.

Abstract

This paper introduces and studies zero-base generalized few-shot learning (zero-base GFSL), which is an extreme yet practical version of few-shot learning problem. Motivated by the cases where base data is not available due to privacy or ethical issues, the goal of zero-base GFSL is to newly incorporate the knowledge of few samples of novel classes into a pretrained model without any samples of base classes. According to our analysis, we discover the fact that both mean and variance of the weight distribution of novel classes are not properly established, compared to those of base classes. The existing GFSL methods attempt to make the weight norms balanced, which we find helps only the variance part, but discard the importance of mean of weights particularly for novel classes, leading to the limited performance in the GFSL problem even with base data. In this paper, we overcome this limitation by proposing a simple yet effective normalization method that can effectively control both mean and variance of the weight distribution of novel classes without using any base samples and thereby achieve a satisfactory performance on both novel and base classes. Our experimental results somewhat surprisingly show that the proposed zero-base GFSL method that does not utilize any base samples even outperforms the existing GFSL methods that make the best use of base data. Our implementation is available at: https://github.com/bigdata-inha/Zero-Base-GFSL.