Miles: Metric Learning with Expandable Subspace for Pre-Trained Model-Based Class-Incremental Learning

2026-07-20

AI Summary

This work addresses catastrophic forgetting and parameter inflation in pre-trained models under class-incremental learning by proposing an expandable subspace mechanism. The approach decouples learnable modules from the pre-trained backbone and leverages intermediate features to guide the construction of task-specific subspaces. By integrating center loss with distance regularization, it simultaneously clusters new classes while preserving cross-task metric consistency. This design enables efficient integration of pre-trained knowledge, achieving both parameter-efficient expansion and stable metric structure. The method attains state-of-the-art performance across six benchmark datasets.

Abstract

Class Incremental Learning (CIL) aims to learn new concepts consistently from a data stream without forgetting. Unlike typical CIL methods which need to learn a model from scratch, pre-trained model (PTM) can easily adapt to a new task with fine-tuning. However, existing PTM-based CIL methods fail to achieve a trade-off between performance and computational expenditure, i.e., they either adopt the same parameter space so that leading catastrophic forgetting, or expand a new branch for each task but adding more computational cost. To this end, we propose MetrIc Learning with Expandable Subspace (Miles) to harness the prior information within pre-trained knowledge, thereby orchestrating an efficient expansion of the parameter space through guided optimization. Specifically, it decouples the learnable modules with the pre-trained model, exploiting prior information from intermediate features of the backbone network to enable more flexible parameter expansion. Then, a central loss is adopted to guide the new category to cluster towards the corresponding prototype in the new task subspace while incorporating an auxiliary distance regularization term to maintain metric equilibrium across tasks. Extensive experiments on six benchmark datasets demonstrate that Miles achieves state-of-the-art performance in various CIL settings.

Problem

• Class-Incremental Learning
• Pre-Trained Model
• Catastrophic Forgetting
• Parameter Expansion
• Computational Efficiency

Innovation

• Class-Incremental Learning
• Pre-Trained Models
• Expandable Subspace
• Metric Learning
• Parameter Expansion