English

Continual Learning on a Data Diet

Machine Learning 2024-10-24 v1 Computer Vision and Pattern Recognition

Abstract

Continual Learning (CL) methods usually learn from all available data. However, this is not the case in human cognition which efficiently focuses on key experiences while disregarding the redundant information. Similarly, not all data points in a dataset have equal potential; some can be more informative than others. This disparity may significantly impact the performance, as both the quality and quantity of samples directly influence the model's generalizability and efficiency. Drawing inspiration from this, we explore the potential of learning from important samples and present an empirical study for evaluating coreset selection techniques in the context of CL to stimulate research in this unexplored area. We train different continual learners on increasing amounts of selected samples and investigate the learning-forgetting dynamics by shedding light on the underlying mechanisms driving their improved stability-plasticity balance. We present several significant observations: learning from selectively chosen samples (i) enhances incremental accuracy, (ii) improves knowledge retention of previous tasks, and (iii) refines learned representations. This analysis contributes to a deeper understanding of selective learning strategies in CL scenarios.

Keywords

Cite

@article{arxiv.2410.17715,
  title  = {Continual Learning on a Data Diet},
  author = {Elif Ceren Gok Yildirim and Murat Onur Yildirim and Joaquin Vanschoren},
  journal= {arXiv preprint arXiv:2410.17715},
  year   = {2024}
}

Comments

18 pages, 6 figures

R2 v1 2026-06-28T19:32:39.362Z