通过再生正则化保持持续学习中的可塑性
Maintaining Plasticity in Continual Learning via Regenerative Regularization
AI总结：
针对持续学习中神经网络可塑性丧失的问题，本文提出 L2 Init，通过朝向初始参数的 L2 正则化使不敏感参数回归初始值，并在多类非平稳监督任务中展现出最稳定的缓解效果。
摘要：
在持续学习中，可塑性是指智能体快速适应新信息的能力。众所周知，神经网络在处理非平稳数据流时会丧失可塑性。本文提出 L2 Init，这是一种通过在损失函数中加入朝向初始参数的 L2 正则化来保持可塑性的简单方法。它与标准 L2 正则化（L2）非常相似，唯一区别在于 L2 是朝向原点进行正则化。L2 Init 实现简单，只需选择一个超参数。该方法的动机与重置神经元或参数值的方法相同。直观而言，当近期损失对某些参数不敏感时，这些参数应向其初始值漂移，从而为快速适应新任务做好准备。在代表持续监督学习中不同类型非平稳性的问题上，我们证明，与以往提出的方法相比，L2 Init 最能持续缓解可塑性损失。
英文摘要：
In continual learning, plasticity refers to the ability of an agent to quickly adapt to new information. Neural networks are known to lose plasticity when processing non-stationary data streams. In this paper, we propose L2 Init, a simple approach for maintaining plasticity by incorporating in the loss function L2 regularization toward initial parameters. This is very similar to standard L2 regularization (L2), the only difference being that L2 regularizes toward the origin. L2 Init is simple to implement and requires selecting only a single hyper-parameter. The motivation for this method is the same as that of methods that reset neurons or parameter values. Intuitively, when recent losses are insensitive to particular parameters, these parameters should drift toward their initial values. This prepares parameters to adapt quickly to new tasks. On problems representative of different types of nonstationarity in continual supervised learning, we demonstrate that L2 Init most consistently mitigates plasticity loss compared to previously proposed approaches.
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。
