Continual learning (CL) is a setting in which an agent has to learn from an
incoming stream of data during its entire lifetime. Although major advances
have been made in the field, one recurring problem which remains unsolved is
that of Catastrophic Forgetting (CF). While the issue has been extensively
studied empirically, little attention has been paid from a theoretical angle.
In this paper, we show that the impact of CF increases as two tasks
increasingly align. We introduce a measure of task similarity called the NTK
overlap matrix which is at the core of CF. We analyze common projected gradient
algorithms and demonstrate how they mitigate forgetting. Then, we propose a
variant of Orthogonal Gradient Descent (OGD) which leverages structure of the
data through Principal Component Analysis (PCA). Experiments support our
theoretical findings and show how our method can help reduce CF on classical CL
datasets.

本文介绍了一个叫做 “持续学习” 的学习模式，研究了一个叫做 “灾难遗忘” 的问题，提出了一种叫做 “NTK overlap matrix” 的任务相似度度量方法，并介绍了一些算法来减轻灾难遗忘产生的影响。其中，我们提出一种基于主成分分析的正交梯度下降算法可以帮助减轻灾难遗忘在常见持续学习场景下的影响。实验结果表明，我们的方法能够有效减轻灾难遗忘的问题。