在持续学习中寻找结构
Finding Structure in Continual Learning
AI总结：
本文提出了一种基于Douglas-Rachford Splitting方法的持续学习框架，通过解耦的两个目标在稳定性和可塑性之间进行协商，实现了更高效且稳定的持续学习。
摘要：
从一系列任务中学习通常面临可塑性与稳定性的矛盾：获取新知识往往导致对过去信息的灾难性遗忘。大多数方法通过求和竞争损失项来解决这一问题，产生梯度冲突，通常需要复杂的且效率低的策略如外部记忆回放或参数正则化来管理。我们提出了一种使用Douglas-Rachford Splitting（DRS）重新表述持续学习目标的方法。这种方法将学习过程重新表述为两个解耦目标之间的协商：一个促进新任务的可塑性，另一个确保旧知识的稳定性。通过迭代地通过其近端算子寻找共识，DRS提供了一种更加系统和稳定的持续学习动态。我们的方法在不需辅助模块或复杂附加组件的情况下实现了稳定性与可塑性之间的高效平衡，为持续学习系统提供了一种更简单却更强大的范式。
英文摘要：
Learning from a stream of tasks usually pits plasticity against stability: acquiring new knowledge often causes catastrophic forgetting of past information. Most methods address this by summing competing loss terms, creating gradient conflicts that are managed with complex and often inefficient strategies such as external memory replay or parameter regularization. We propose a reformulation of the continual learning objective using Douglas-Rachford Splitting (DRS). This reframes the learning process not as a direct trade-off, but as a negotiation between two decoupled objectives: one promoting plasticity for new tasks and the other enforcing stability of old knowledge. By iteratively finding a consensus through their proximal operators, DRS provides a more principled and stable learning dynamic. Our approach achieves an efficient balance between stability and plasticity without the need for auxiliary modules or complex add-ons, providing a simpler yet more powerful paradigm for continual learning systems.
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。
