Miles: Metric Learning with Expandable Subspace for Pre-Trained Model-Based Class-Incremental Learning
2026-07-20 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors focus on a problem called Class Incremental Learning, where a model must learn new things without forgetting older knowledge. They note that using pre-trained models can help but existing methods either forget old information or get too big and slow. Their solution, called Miles, cleverly uses parts of the pre-trained model to add new knowledge more efficiently without growing too large. They also use techniques to keep new information organized and balanced with old information. Tests show their approach works better than others on several datasets.
Class Incremental LearningPre-trained ModelFine-tuningCatastrophic ForgettingParameter ExpansionMetric LearningPrototypeLoss FunctionBackbone NetworkRegularization
Authors
Kai Jiang, Zisong Lin, Hongyuan Zhang, Xueru Bai, Xuelong Li
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.