Class-Incremental Learning (CIL) aims to enable models to sequentially learn new tasks while retaining knowledge from previous ones. Recently, merging-based pre-trained CIL methods have gained significant attention due to their competitive performance and high inference efficiency. However, most existing approaches decouple training from merging, neglecting the compatibility among task-specific adapters. This incompatibility introduces severe conflicts during integration, resulting in catastrophic forgetting and notable performance degradation. To address this limitation, we propose MOBO, a Merging-Oriented Bi-Level Optimization framework that synergizes the optimization of the current task model with the performance of the final merged model. At the upper level, we introduce a global loss that anticipates the merged model's behavior, guiding the current task parameters toward a solution space that facilitates effective merging. At the lower level, we employ a dynamic weighted merging strategy to optimize merging coefficients and update the merged model. By alternately optimizing the task-specific and merged models, MOBO effectively mitigates the performance loss caused by incompatible adapter integration. Comprehensive experiments on CIFAR-100, CUB-200, ImageNet-R, ImageNet-A, and VTAB demonstrate the superiority of our approach, highlighting its robustness and scalability, particularly on long task sequences.
Si-Yu Zhang, Wen Wang, Wen-Ju Sun et al.· Proceedings of the Thirty-Fi...· 0 citations
Incremental learning (IL) aims to continually acquire new knowledge (plasticity) while retaining previously learned information (stability). However, striking a balance between plasticity and stability remains a significant challenge for intelligent systems. The human brain achieves exceptional balance, owing to various memory units that collaboratively encode and store information. Inspired by human memory mechanisms, this paper introduces an IL framework with knowledge enCoding, Consolidation, and Integration (Co $^{2}$ In). Co $^{2}$ In is designed as a bi-level memory architecture: a working memory for adaptive knowledge acquisition and a long-term memory dedicated to the persistent retention of information. The two memory modules work cooperatively during IL. The working memory first learns from new data to encode knowledge into parameters. Subsequently, Co $^{2}$ In performs a consolidation process to identify underlying patterns in the learned parameters and re-express them into a compact knowledge representation. Next, the knowledge representation and the identified patterns are transformed into separate network layers and integrated into the long-term memory. These designs empower Co $^{2}$ In to accumulate knowledge with high plasticity and stability. We evaluate Co $^{2}$ In on CIFAR-10, CIFAR-100, and Tiny-ImageNet under exemplar-free Class-IL and Task-IL settings. Experimental results show that Co $^{2}$ In achieves state-of-the-art performance with efficient memory consumption.
Wenju Sun, Qingyong Li, Boyang Li et al.· IEEE Transactions on Pattern...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
