DualCP: Rehearsal-Free Domain-Incremental Learning via Dual-Level Concept Prototype
Qiang Wang, Yuhang He, Songlin Dong, Xiang Song, Jizhou Han, Haoyu Luo, Yihong Gong
Abstract
Domain-Incremental Learning (DIL) enables vision models to adapt to changing conditions in real-world environments while maintaining the knowledge acquired from previous domains. Given privacy concerns and training time, Rehearsal-Free DIL (RFDIL) is more practical. Inspired by the incremental cognitive process of the human brain, we design Dual-level Concept Prototypes (DualCP) for each class to address the conflict between learning new knowledge and retaining old knowledge in RFDIL. To construct DualCP, we propose a Concept Prototype Generator (CPG) that generates both coarse-grained and fine-grained prototypes for each class. Additionally, we introduce a Coarse-to-Fine calibrator (C2F) to align image features with DualCP. Finally, we propose a Dual Dot-Regression (DDR) loss function to optimize our C2F module. Extensive experiments on the DomainNet, CDDB, and CORe50 datasets demonstrate the effectiveness of our method.
BibTeX
@article{Wang_He_Dong_Song_Han_Luo_Gong_2025, title={DualCP: Rehearsal-Free Domain-Incremental Learning via Dual-Level Concept Prototype}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/35418}, DOI={10.1609/aaai.v39i20.35418}, abstractNote={Domain-Incremental Learning (DIL) enables vision models to adapt to changing conditions in real-world environments while maintaining the knowledge acquired from previous domains. Given privacy concerns and training time, Rehearsal-Free DIL (RFDIL) is more practical. Inspired by the incremental cognitive process of the human brain, we design Dual-level Concept Prototypes (DualCP) for each class to address the conflict between learning new knowledge and retaining old knowledge in RFDIL. To construct DualCP, we propose a Concept Prototype Generator (CPG) that generates both coarse-grained and fine-grained prototypes for each class. Additionally, we introduce a Coarse-to-Fine calibrator (C2F) to align image features with DualCP. Finally, we propose a Dual Dot-Regression (DDR) loss function to optimize our C2F module. Extensive experiments on the DomainNet, CDDB, and CORe50 datasets demonstrate the effectiveness of our method.}, number={20}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Wang, Qiang and He, Yuhang and Dong, Songlin and Song, Xiang and Han, Jizhou and Luo, Haoyu and Gong, Yihong}, year={2025}, month={Apr.}, pages={21198-21206} }