Federated Few-Shot Class-Incremental Learning
Abstract
This study proposes a challenging yet practical Federated Few-Shot Class-Incremental Learning (FFSCIL) problem, where clients only hold very few samples for new classes. We develop a novel Unified Optimized Prototype Prompt (UOPP) model to simultaneously handle catastrophic forgetting, over-fitting, and prototype bias in FFSCIL. UOPP utilizes task-wise prompt learning to mitigate task interference and over-fitting, unified static-dynamic prototypes to achieve a stability-plasticity balance, and adaptive dual heads for enhanced inferences. Dynamic prototypes represent new classes in the current few-shot task and are rectified to deal with prototype bias. Our comprehensive experimental results show that UOPP significantly outperforms state-of-the-art (SOTA) methods on three datasets with improvements up to 76% on average accuracy and 90% on harmonic mean accuracy respectively. Our extensive analysis shows UOPP robustness in various numbers of local clients and global rounds, low communication costs, and moderate running time. The source code of UOPP is publicly available at https://github.com/anwarmaxsum/FFSCIL.
Cite
Text
Ma'sum et al. "Federated Few-Shot Class-Incremental Learning." International Conference on Learning Representations, 2025.Markdown
[Ma'sum et al. "Federated Few-Shot Class-Incremental Learning." International Conference on Learning Representations, 2025.](https://mlanthology.org/iclr/2025/masum2025iclr-federated/)BibTeX
@inproceedings{masum2025iclr-federated,
title = {{Federated Few-Shot Class-Incremental Learning}},
author = {Ma'sum, Muhammad Anwar and Pratama, Mahardhika and Liu, Lin and Habibullah, H and Kowalczyk, Ryszard},
booktitle = {International Conference on Learning Representations},
year = {2025},
url = {https://mlanthology.org/iclr/2025/masum2025iclr-federated/}
}