Continual learning aims to learn on non-stationary data streams without catastrophically forgetting previous knowledge. Prevalent replay-based methods address this challenge by rehearsing on a small buffer holding the seen data, for which a delicate sample selection strategy is required. However, existing selection schemes typically seek only to maximize the utility of the ongoing selection, overlooking the interference between successive rounds of selection. Motivated by this, we dissect the interaction of sequential selection steps within a framework built on influence functions. We manage to identify a new class of second-order influences that will gradually amplify incidental bias in the replay buffer and compromise the selection process. To regularize the second-order effects, a novel selection objective is proposed, which also has clear connections to two widely adopted criteria. Furthermore, we present an efficient implementation for optimizing the proposed criterion. Experiments on multiple continual learning benchmarks demonstrate the advantage of our approach over state-of-the-art methods. Code is available at https://github.com/feifeiobama/InfluenceCL.

翻译：持续学习旨在非平稳数据流上进行学习，同时避免灾难性地遗忘先前知识。主流的基于回放的方法通过在小缓冲区内重放已见数据来解决这一挑战，这需要精心设计的样本选择策略。然而，现有选择方案通常仅追求最大化当前选择的效用，忽略了连续选择轮次之间的相互干扰。受此启发，我们在基于影响函数构建的框架内剖析了顺序选择步骤的交互作用。我们成功识别出一类新的二阶影响，这些影响会逐步放大回放缓冲区中的偶然偏差，并损害选择过程。为对二阶效应进行正则化，我们提出了一种新颖的选择目标函数，该函数与两种广泛采用的标准具有明确联系。此外，我们还提出了一种优化该标准的高效实现方法。在多个持续学习基准上的实验表明，我们的方法优于现有最先进技术。代码请见 https://github.com/feifeiobama/InfluenceCL。