In this work, we study the problem of continual learning (CL) where the goal
is to learn a model on a sequence of tasks, such that the data from the
previous tasks becomes unavailable while learning on the current task data. CL
is essentially a balancing act between being able to learn on the new task
(i.e., plasticity) and maintaining the performance on the previously learned
concepts (i.e., stability). With an aim to address the stability-plasticity
trade-off, we propose to perform weight-ensembling of the model parameters of
the previous and current task. This weight-ensembled model, which we call
Continual Model Averaging (or CoMA), attains high accuracy on the current task
by leveraging plasticity, while not deviating too far from the previous weight
configuration, ensuring stability. We also propose an improved variant of CoMA,
named Continual Fisher-weighted Model Averaging (or CoFiMA), that selectively
weighs each parameter in the weight ensemble by leveraging the Fisher
information of the weights of the model. Both the variants are conceptually
simple, easy to implement, and effective in attaining state-of-the-art
performance on several standard CL benchmarks.

在这项研究中，我们研究了增量学习（CL）的问题，其目标是在一系列任务中学习模型，使得先前任务的数据在学习当前任务数据时不可用。我们提出了一种称为 Continual Model Averaging（或 CoMA）的加权集成模型，它在保持稳定性的同时，通过利用可塑性，在当前任务上达到高准确性。我们还提出了一种改进的 CoMA 变体，称为 Continual Fisher-weighted Model Averaging（或 CoFiMA），通过利用模型权重的 Fisher 信息，有选择地对加权集合中的每个参数进行加权。这两种变体概念简单、易于实现，并在多个标准 CL 基准测试中实现了最先进的性能。