Few-shot class-incremental learning (FSCIL) struggles to incrementally recognize novel classes from few examples without catastrophic forgetting of old classes or overfitting to new classes. We propose TLCE, which ensembles multiple pre-trained models to improve separation of novel and old classes. TLCE minimizes interference between old and new classes by mapping old class images to quasi-orthogonal prototypes using episodic training. It then ensembles diverse pre-trained models to better adapt to novel classes despite data imbalance. Extensive experiments on various datasets demonstrate that our transfer learning ensemble approach outperforms state-of-the-art FSCIL methods.
@article{arxiv.2312.04225,
title = {TLCE: Transfer-Learning Based Classifier Ensembles for Few-Shot Class-Incremental Learning},
author = {Shuangmei Wang and Yang Cao and Tieru Wu},
journal= {arXiv preprint arXiv:2312.04225},
year = {2023}
}