A Unified Approach to Domain Incremental Learning with Memory: Theory and Algorithm
This work addresses the challenge of adapting to sequential domains with limited memory for practitioners in machine learning, but it is incremental as it builds on and unifies existing methods.
The paper tackles the problem of domain incremental learning by proposing a unified framework (UDIL) that integrates various existing methods and achieves a tighter generalization error bound, with empirical results showing it outperforms state-of-the-art methods on synthetic and real-world datasets.
Domain incremental learning aims to adapt to a sequence of domains with access to only a small subset of data (i.e., memory) from previous domains. Various methods have been proposed for this problem, but it is still unclear how they are related and when practitioners should choose one method over another. In response, we propose a unified framework, dubbed Unified Domain Incremental Learning (UDIL), for domain incremental learning with memory. Our UDIL **unifies** various existing methods, and our theoretical analysis shows that UDIL always achieves a tighter generalization error bound compared to these methods. The key insight is that different existing methods correspond to our bound with different **fixed** coefficients; based on insights from this unification, our UDIL allows **adaptive** coefficients during training, thereby always achieving the tightest bound. Empirical results show that our UDIL outperforms the state-of-the-art domain incremental learning methods on both synthetic and real-world datasets. Code will be available at https://github.com/Wang-ML-Lab/unified-continual-learning.