Skip to main navigation menu Skip to main content Skip to site footer

Vol. 12 No. 10 (2025)
International Multidisciplinary Journal for Research & Development

Articles

SYNAPTIC PLASTICITY AND CRITICAL LEARNING PERIODS IN DEEP CONVOLUTIONAL NETWORKS: BRIDGING BIOLOGICAL MECHANISMS WITH MACHINE REPRESENTATION

Published 2025-10-31

  • Dr. Elias V. Thorne
    • ,
  • Dr. Sarah J. Mendez

Dr. Elias V. Thorne
Department of Computational Neuroscience, Institute of Advanced Systems

Dr. Sarah J. Mendez
Department of Computational Neuroscience, Institute of Advanced Systems

Abstract

The Background: Deep Convolutional Neural Networks (CNNs) have achieved remarkable success in static classification tasks. However, unlike biological neural systems, they often struggle with continuous learning and adaptability, prone to catastrophic forgetting when introduced to new data distributions.

Methods: This study proposes a hybrid architecture, the "Plastic-CNN" (P-CNN), which integrates principles of synaptic plasticity and adult neurogenesis into standard deep learning frameworks. We compare the performance of standard architectures (VGG-16, AlexNet) against the P-CNN using the iNaturalist 2017 dataset for fine-grained visual classification and a financial dataset for dynamic fraud detection. We further analyze the role of "critical learning periods" by modulating information bottlenecks during the initial training epochs.

Results: The P-CNN demonstrated a statistically significant improvement in long-term feature retention compared to static baselines. Specifically, the inclusion of neurogenic node-addition layers reduced validation loss during domain shifts. The analysis confirms that applying information constraints during the early "critical period" of training creates more robust generalized representations, mirroring biological sensory development.

Conclusion: Incorporating biological constraints such as synaptic plasticity and neurogenesis does not merely mimic the brain but offers a tangible computational advantage for addressing the stability-plasticity dilemma in artificial intelligence. These findings suggest that future architectures should prioritize dynamic topology over static depth.

Keywords

Deep Learning, Synaptic Plasticity, Convolutional Neural Networks, Neurogenesis

pdf

References

  1. Dip Bharatbhai Patel. (2025). Comparing Neural Networks and Traditional Algorithms in Fraud Detection. The American Journal of Applied Sciences, 7(07), 128–132. https://doi.org/10.37547/tajas/Volume07Issue07-13
  2. Jiuxiang Gu, Zhenhua Wang, Jason Kuen, Lianyang Ma, Amir Shahroudy, Bing Shuai, Ting Liu, Xingxing Wang, and Gang Wang. Recent advances in convolutional neural networks. arXiv preprint arXiv:1512.07108, 2015.
  3. Matthew D Zeiler and Rob Fergus. Visualizing and understanding convolutional networks. In European conference on computer vision, pages 818–833. Springer, 2014.
  4. Yann LeCun, Bernhard Boser, John S Denker, Donnie Henderson, Richard E Howard, Wayne Hubbard, and Lawrence D Jackel. Backpropagation applied to handwritten zip code recognition. Neural computation, 1(4):541–551, 1989.
  5. Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems, pages 1097–1105, 2012.
  6. Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.
  7. Grant Van Horn, Oisin Mac Aodha, Yang Song, Alex Shepard, Hartwig Adam, Pietro Perona, and Serge Belongie. The inaturalist challenge 2017 dataset. arXiv preprint arXiv:1707.06642, 2017.
  8. Abbott, L. F. & Nelson, S. B. (2000), ‘Synaptic plasticity: taming the beast’, Nature Neuroscience 3, 1178–1183.
  9. Abraham, W. C. & Robins, A. (2005), ‘Memory retention and weight plasticity in ANN simulations’, Trends in Neurosciences 28(2), 73–78.
  10. Achille, A., Rovere, M. & Soatto, S. (2017), Critical learning periods in deep neural networks, arXiv:1711.08856.
  11. Adams, R. A., Friston, K. J. & Bastos, A. M. (2015), ‘Active inference, predictive coding and cortical architecture’, In M. F. Casanova and I. Opris (Eds.), Recent Advances on the Modular Organization of the Cortex. Springer Netherlands.
  12. Aimone, J. B., Wiles, J. & Gage, F. H. (2006), ‘Potential role for adult neurogenesis in the encoding of time in new memories.’, Nature Neuroscience 9, 723–727.
  13. Aimone, J. B., Wiles, J. & Gage, F. H. (2009), ‘Computational influence of adult neurogenesis on memory encoding.’, Neuron 61, 187–202.
  14. Altman, J. (1963), ‘Autoradiographic investigation of cell proliferation in the brains of rats and cats’, The Anatomical Record 145(4), 573–591.
  15. Astrom, K. J. & Murray, R. M. (2010), Feedback Systems: An Introduction for Scientists and Engineers, Princeton University Press

Downloads

Download data is not yet available.