Divide and Conquer: Static-Dynamic Collaboration for Few-Shot Class-Incremental Learning

TL;DR

This paper introduces a two-stage framework called Static-Dynamic Collaboration for few-shot class-incremental learning, effectively balancing the retention of old knowledge with learning new classes, and achieves state-of-the-art results.

Contribution

The paper proposes dividing FSCIL into static retaining and dynamic learning stages, with a novel static-dynamic collaboration framework for improved stability-plasticity trade-off.

Findings

Achieves state-of-the-art performance on three benchmarks.

Effectively balances old knowledge retention and new class adaptation.

Demonstrates robustness in real-world application dataset.

Abstract

Few-shot class-incremental learning (FSCIL) aims to continuously recognize novel classes under limited data, which suffers from the key stability-plasticity dilemma: balancing the retention of old knowledge with the acquisition of new knowledge. To address this issue, we divide the task into two different stages and propose a framework termed Static-Dynamic Collaboration (SDC) to achieve a better trade-off between stability and plasticity. Specifically, our method divides the normal pipeline of FSCIL into Static Retaining Stage (SRS) and Dynamic Learning Stage (DLS), which harnesses old static and incremental dynamic class information, respectively. During SRS, we train an initial model with sufficient data in the base session and preserve the key part as static memory to retain fundamental old knowledge. During DLS, we introduce an extra dynamic projector jointly trained with the previous static memory. By employing both stages, our method achieves improved retention of old knowledge while continuously adapting to new classes. Extensive experiments on three public benchmarks and a real-world application dataset demonstrate that our method achieves state-of-the-art performance against other competitors.