Spiking Neural Networks (SNNs) offer a biologically plausible and energy-efficient paradigm for processing temporal data, particularly suited for neuromorphic computing platforms. However, their deterministic nature limits their deployment in safety-critical applications where reliable uncertainty quantification is essential. This paper introduces a novel Bayesian Spiking Neural Network (BSNN) framework that integrates variational inference with surrogate gradient learning to enable robust uncertainty estimation while preserving the efficiency of SNNs. We formulate a scalable Bayesian framework using mean-field variational approximation over network weights and biases, enabling full predictive uncertainty quantification through Monte Carlo sampling. To address the non-differentiability of spiking dynamics, we employ a smooth surrogate gradient method based on sigmoidal derivatives during backpropagation through time. A tailored Poisson encoding scheme ensures rich temporal input representation, while a KL annealing strategy stabilizes training by gradually increasing the regularization pressure from prior distributions. Comprehensive experiments on a synthetic dataset demonstrate competitive classification accuracy (77.78%) alongside well-calibrated uncertainty estimates, as evidenced by strong calibration curves and high AUROC scores (> 0.9). The model exhibits stable training dynamics, with controlled gradients and balanced loss components. Our approach bridges the gap between efficient neuromorphic computation and trustworthy AI, offering a pathway toward deployable, uncertainty-aware edge intelligence systems.
Banteywalu et al. (Thu,) studied this question.