EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Continual Learning for Streaming Data with Dynamic Memory Allocation and Drift-Aware Rehearsal

Yan Lu, Alessandro Rossi

International Journal of Computational and Biological Sciences · 31.12.2025 · 2 (02) · 221-230

DOI: 10.66238/ijcbs27 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computational and Biological Sciences, online erschienen am 31.12.2025. Verfasst von Yan Lu und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The capability of artificial neural networks to learn continuously from non-stationary data streams without catastrophic forgetting remains a paramount challenge in machine learning. While Experience Replay (ER) has emerged as a robust strategy, standard approaches often rely on fixed-size memory buffers and uniform sampling, which are suboptimal for streaming data exhibiting significant concept drift and class imbalance. In this paper, we propose a novel framework titled Dynamic Memory Allocation with Drift-Aware Rehearsal (DM-DAR). Our approach introduces a dynamic budgeting mechanism that adjusts the memory quota per class based on real-time estimation of forgetting and feature space drift. Furthermore, we implement a dual-criterion retrieval strategy that prioritizes samples located near the decision boundary and those representing the centroid of drifting distributions. Extensive experiments on split-CIFAR100 and sequential TinyImage Net benchmarks demonstrate that DM-DAR significantly outperforms state-of-the-art baselines in terms of average accuracy and backward transfer, while maintaining computational efficiency. The results suggest that adaptive resource management is crucial for the longevity and plasticity of continual learning systems deployed in volatile environments.

Zitieren

APA

Lu, Y., & Rossi, A. (2025). Continual Learning for Streaming Data with Dynamic Memory Allocation and Drift-Aware Rehearsal. International Journal of Computational and Biological Sciences, 2(02), 221-230. https://doi.org/10.66238/ijcbs27

Vancouver

Lu Y, Rossi A. Continual Learning for Streaming Data with Dynamic Memory Allocation and Drift-Aware Rehearsal. International Journal of Computational and Biological Sciences. 2025;2(02):221-230. doi:10.66238/ijcbs27

BibTeX
@article{Lu2025,
  author = {Lu, Yan and Rossi, Alessandro},
  title = {Continual Learning for Streaming Data with Dynamic Memory Allocation and Drift-Aware Rehearsal},
  journal = {International Journal of Computational and Biological Sciences},
  year = {2025},
  volume = {2},
  number = {02},
  pages = {221--230},
  doi = {10.66238/ijcbs27}
}
Kostenlos starten →