Achieving Optimal Accuracy and Robustness Through Tight Excitatory–Inhibitory Balance in Shallow Spiking Recurrent Neural Network
International Journal of Neural Systems · 2026
DOI: 10.1142/s0129065726500176 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Neural Systems, 2026. Verfasst von Shiwen Li und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Artificial Intelligence, Shenzhen Technology University, Shenzhen 518118, P. R. China. Hauptthemen laut MEDLINE: Neural Inhibition, Neurons, Action Potentials, Models, Neurological. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 81 Literaturangaben. Gefördert von: Shenzhen Science and Technology Program, Shenzhen Science and Technology Major Project, Major Project of Science and Technology Research Program of Chongqing Education Commission.
Abstract (Originalsprache)
Traditional deep neural networks exhibit high computational complexity during training and lack biological interpretability due to their reliance on backpropagation-based methods. Spiking Recurrent Neural Network (SRNN) performs well in processing spatio-temporal information by using discrete spike events. It attracts increasing attention in neural computing due to its biological plausibility and hardware implementation. To improve the performance of SRNN, we propose an excitation–inhibition balanced shallow SRNN (EI-SRNN), which is inspired by the balance of excitation and inhibition in the brain, by optimizing the input currents of reservoir neurons to achieve a tight balanced state. The proposed EI-SRNN achieves optimal accuracy while maintaining low computational complexity, debunking the conventional trade-off between accuracy and robustness. We analyze the neural encoding ability and information memory capacity of the EI-SRNN and compare the performance of the model under different degrees of excitation and inhibition. Our experiments demonstrate that EI-SRNN can have higher neural coding capacity and memory capacity under tight balanced excitatory and inhibitory balanced states, so it can achieve better accuracy while possessing stronger robustness. Furthermore, when the reservoir is dominated by excitatory influences, performance declines faster than when the reservoir is dominated by inhibitory influences.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Artificial Intelligence, Shenzhen Technology University, Shenzhen 518118, P. R. China.
- Faculty of Data Science, City University of Macau, Macau 999078, P. R. China.
- School of Artificial Intelligence, Xidian University, Xi'an, Shaanxi 710071, P. R. China.
- School of Mathematics and Computer Science, Chongqing College of International Business and Economics, No. 88 Xuefu Road, Hechuan District, Chongqing, P. R. China.
ORCID-Profile: Shiwen Li, Junsong Wang, Syeda Shamaila Zareen
Förderung
- Shenzhen Science and Technology Program
- Shenzhen Science and Technology Major Project
- Major Project of Science and Technology Research Program of Chongqing Education Commission
- Shenzhen University of Technology Self-made Experimental Instruments and Equipment Project
- Open Fund of National Engineering Laboratory for Big Data System Computing Technology
- Guangxi Key Laboratory of Brain-inspired Computing and Intelligent Chips
- National Natural Science Foundation of China
Zitieren
APA
Li, S., Wang, J., & Zareen, S. S. (2026). Achieving Optimal Accuracy and Robustness Through Tight Excitatory–Inhibitory Balance in Shallow Spiking Recurrent Neural Network. International Journal of Neural Systems. https://doi.org/10.1142/s0129065726500176
Vancouver
Li S, Wang J, Zareen SS. Achieving Optimal Accuracy and Robustness Through Tight Excitatory–Inhibitory Balance in Shallow Spiking Recurrent Neural Network. International Journal of Neural Systems. 2026. doi:10.1142/s0129065726500176
BibTeX
@article{Li2026,
author = {Li, Shiwen and Wang, Junsong and Zareen, Syeda Shamaila},
title = {Achieving Optimal Accuracy and Robustness Through Tight Excitatory–Inhibitory Balance in Shallow Spiking Recurrent Neural Network},
journal = {International Journal of Neural Systems},
year = {2026},
doi = {10.1142/s0129065726500176}
}
EUVIMED