Continual learning is a promising machine learning paradigm to learn new tasks while retaining previously learned knowledge over streaming training data. Till now, rehearsal-based methods, keeping a small part of data from old tasks as a memory buffer, have shown good performance in mitigating catastrophic forgetting for previously learned knowledge. However, most of these methods typically treat each new task equally, which may not adequately consider the relationship or similarity between old and new tasks. Furthermore, these methods commonly neglect sample importance in the continual training process and result in sub-optimal performance on certain tasks. To address this challenging problem, we propose Relational Experience Replay (RER), a bi-level learning framework, to adaptively tune task-wise relationships and sample importance within each task to achieve a better `stability' and `plasticity' trade-off. As such, the proposed method is capable of accumulating new knowledge while consolidating previously learned old knowledge during continual learning. Extensive experiments conducted on three publicly available datasets (i.e., CIFAR-10, CIFAR-100, and Tiny ImageNet) show that the proposed method can consistently improve the performance of all baselines and surpass current state-of-the-art methods.

翻译：持续学习是一种有前景的机器学习范式，旨在从流式训练数据中学习新任务的同时保留先前学到的知识。迄今为止，基于回忆的方法通过保留少量旧任务数据作为记忆缓冲区，在缓解先前所学知识的灾难性遗忘方面表现良好。然而，这些方法大多通常平等对待每个新任务，可能未充分考虑新旧任务之间的关系或相似性。此外，这些方法普遍忽略持续训练过程中的样本重要性，导致在某些任务上表现次优。为解决这一具有挑战性的问题，我们提出关系经验回放（RER），一种双层学习框架，用于自适应调整任务间关系及每个任务内的样本重要性，以实现更好的“稳定性”与“可塑性”权衡。由此，所提方法能够在持续学习过程中积累新知识的同时巩固先前学到的旧知识。在三个公开可用数据集（即CIFAR-10、CIFAR-100和Tiny ImageNet）上开展的大量实验表明，所提方法能够持续提升所有基线的性能，并超越当前最先进的方法。