Enhancing Few-Shot Classification without Forgetting through Multi-Level Contrastive Constraints

TL;DR

This paper introduces MLCC, a novel framework that enhances few-shot classification by jointly addressing inductive bias and catastrophic forgetting through multi-level contrastive constraints and distribution alignment.

Contribution

The paper proposes a unified learning paradigm with space-aware interaction modeling and cross-stage distribution adaptation to improve few-shot learning performance.

Findings

MLCC outperforms state-of-the-art methods on multiple datasets.

It effectively reduces semantic gaps between past and current predictions.

The approach demonstrates consistent improvements across various benchmarks.

Abstract

Most recent few-shot learning approaches are based on meta-learning with episodic training. However, prior studies encounter two crucial problems: (1) \textit{the presence of inductive bias}, and (2) \textit{the occurrence of catastrophic forgetting}. In this paper, we propose a novel Multi-Level Contrastive Constraints (MLCC) framework, that jointly integrates within-episode learning and across-episode learning into a unified interactive learning paradigm to solve these issues. Specifically, we employ a space-aware interaction modeling scheme to explore the correct inductive paradigms for each class between within-episode similarity/dis-similarity distributions. Additionally, with the aim of better utilizing former prior knowledge, a cross-stage distribution adaption strategy is designed to align the across-episode distributions from different time stages, thus reducing the semantic gap between existing and past prediction distribution. Extensive experiments on multiple few-shot datasets demonstrate the consistent superiority of MLCC approach over the existing state-of-the-art baselines.