Fresh-CL: Feature Realignment through Experts on Hypersphere in Continual Learning

TL;DR

Fresh-CL introduces a novel feature realignment method using hypersphere classifiers and mixture of experts to improve task separation in continual learning, demonstrating significant accuracy gains across multiple datasets.

Contribution

The paper proposes a dynamic ETF-based approach combined with mixture of experts to enhance feature separation and mitigate feature entanglement in continual learning.

Findings

Achieves 2% higher accuracy than baseline methods.

Improves feature distinction especially in fine-grained datasets.

Demonstrates effectiveness across 11 diverse datasets.

Abstract

Continual Learning enables models to learn and adapt to new tasks while retaining prior knowledge. Introducing new tasks, however, can naturally lead to feature entanglement across tasks, limiting the model's capability to distinguish between new domain data. In this work, we propose a method called Feature Realignment through Experts on hyperSpHere in Continual Learning (Fresh-CL). By leveraging predefined and fixed simplex equiangular tight frame (ETF) classifiers on a hypersphere, our model improves feature separation both intra and inter tasks. However, the projection to a simplex ETF shifts with new tasks, disrupting structured feature representation of previous tasks and degrading performance. Therefore, we propose a dynamic extension of ETF through mixture of experts, enabling adaptive projections onto diverse subspaces to enhance feature representation. Experiments on 11 datasets demonstrate a 2% improvement in accuracy compared to the strongest baseline, particularly in fine-grained datasets, confirming the efficacy of combining ETF and MoE to improve feature distinction in continual learning scenarios.