A Comparison of Data Selection Methods for Replay Buffer in Continual Learning

Abstract

Continual learning aims to learn a sequence of tasks without forgetting the previously learned tasks. One way to achieve continual learning is by applying a replay buffer that stores previously learned samples to existing learning frameworks. Among these methods, Lifelong Unsupervised Mixup (LUMP) interpolates the current task samples and samples from the replay buffer to prevent forgetting. LUMP randomly selects data for the replay buffer. This random selection is likely to cause a biased data dis-(breakpoint)tribution in the buffer, which has a negative effect on learning performance. In this paper, we propose a clustering-based data selection method for the replay buffer. Additionally, we compare several clustering methods because the effect on learning performance is presumed to depend on clustering methods.