Enhancing Context-Dependent Learning in Spiking Neural Networks Through Symmetric STDP and Synaptic Scaling

Spiking neural networks (SNNs) are heralded for their biological plausibility and efficiency in processing spatiotemporal data. However, maintaining robust learning in noisy environments remains a significant challenge. This paper introduces a novel model that integrates symmetric spike-timing-dependent plasticity (sym-STDP) with synaptic scaling mechanisms to enhance context-dependent learning in SNNs. By dynamically adjusting the synaptic weights and incorporating synaptic scaling, our model achieves superior learning accuracy and robustness to noise. Experimental results demonstrate that our approach significantly outperforms traditional STDP models, achieving higher learning efficiency and adaptability in varying conditions. These findings underscore the potential of our model for advancing the application of SNNs in real-world, dynamic environments.