关系经验回放：通过自适应调节任务间关系进行持续学习
Relational Experience Replay: Continual Learning by Adaptively Tuning Task-wise Relationship
AI总结：
提出关系经验回放（RER）双层学习框架，通过自适应调节任务间关系与样本重要性优化稳定性与可塑性权衡，有效缓解持续学习中的灾难性遗忘，在多个数据集上超越现有方法。
摘要：
持续学习是一种有前景的机器学习范式，旨在流式训练数据上学习新任务的同时保留已学知识。至今，基于排练的方法通过将旧任务的一小部分数据作为记忆缓冲区保留，在缓解已学知识的灾难性遗忘方面表现出良好性能。然而，这些方法通常平等对待每个新任务，可能未充分考虑旧任务与新任务之间的关系或相似性。此外，这些方法通常忽略持续训练过程中的样本重要性，导致在某些任务上性能次优。为解决这一挑战性问题，我们提出关系经验回放（RER），一种双层学习框架，自适应调节任务间关系和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。因此，该方法能够在持续学习过程中积累新知识的同时巩固已学的旧知识。在三个公开数据集（即CIFAR-10、CIFAR-100和Tiny ImageNet）上进行的广泛实验表明，该方法能持续提升所有基线的性能，并超越当前最先进方法。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
