Stochastic engrams for efficient continual learning
APL Machine Learning · 01.12.2026 · 4 (4)
DOI: 10.1063/5.0346731 ↗Kurz gefasst
Fachartikel in der Zeitschrift APL Machine Learning, online erschienen am 06.11.2026. Verfasst von Isabelle Aguilar und 2 weiteren Autor:innen. Die Arbeit nennt 81 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Australian Research Council, Department of Education, Australian Government, University of Sydney.
Abstract (Originalsprache)
The ability to learn continuously in artificial neural networks is often limited by catastrophic forgetting, a phenomenon in which new knowledge becomes dominant. By taking mechanisms of memory encoding in neuroscience (i.e., engrams) as inspiration, we propose a novel approach that integrates stochastically activated engrams as a gating mechanism for metaplastic binarized neural networks (mBNNs). This method leverages the computational efficiency of mBNNs combined with the robustness of probabilistic memory traces to mitigate forgetting and maintain the model’s reliability. Previously validated metaplastic optimization techniques have been incorporated to further enhance synaptic stability. Compared to baseline binarized models and benchmark fully connected continual learning approaches, our method is the only strategy capable of achieving average accuracies over 70% in both class-incremental and domain-incremental MNIST benchmarks, matching full-precision state-of-the-art methods. Furthermore, we achieve a significant reduction in peak GPU and RAM usage, under 5% and 20%, respectively, as well as an ∼8× reduction in memory footprint compared to full precision counterparts. Our findings demonstrate (A) an improved stability vs plasticity trade-off, (B) reduced memory intensiveness, and (C) enhanced performance in binarized architectures. By uniting principles of neuroscience and efficient computing, we offer new insights into the design of scalable and robust deep learning systems.
Förderung
- Australian Research Council
- Department of Education, Australian Government
- University of Sydney
Zitieren
APA
Aguilar, I., Contreras, L. F. H., & Kavehei, O. (2026). Stochastic engrams for efficient continual learning. APL Machine Learning, 4(4). https://doi.org/10.1063/5.0346731
Vancouver
Aguilar I, Contreras LFH, Kavehei O. Stochastic engrams for efficient continual learning. APL Machine Learning. 2026;4(4). doi:10.1063/5.0346731
BibTeX
@article{Aguilar2026,
author = {Aguilar, Isabelle and Contreras, Luis Fernando Herbozo and Kavehei, Omid},
title = {Stochastic engrams for efficient continual learning},
journal = {APL Machine Learning},
year = {2026},
volume = {4},
number = {4},
doi = {10.1063/5.0346731}
}
EUVIMED