SpikingDynamicMaskFormer: Enhancing Efficiency in Spiking Neural Networks with Dynamic Masking
Open Access
- 31 December 2025
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 15 (1) , 189
Abstract
Spiking Neural Networks (SNNs) offer promising low-power alternatives to conventional neural models but often incur considerable redundancy in parameters and computations. To address these inefficiencies, we propose SpikingDynamicMaskFormer (SDMFormer), a novel framework that integrates dynamic masking and lightweight position encoding into a spike-based Transformer backbone. Specifically, our Dynamic Mask Encoder Block adaptively suppresses ineffective spike channels by learning mask parameters, reducing parameter count to 37.93–42.69% of the original Spikformer. Simultaneously, a redesigned lightweight position embedding replaces resource-intensive relative position convolutions, further lowering complexity. Experiments on three neuromorphic vision datasets—DVS128, CIFAR10-DVS and N-Caltech101—demonstrate that SDMFormer cuts energy consumption by 42.79–50.13% relative to Spikformer while maintaining or slightly surpassing accuracy. Moreover, compared with recent leading works, SDMFormer achieves competitive accuracy with substantially fewer parameters and delivers higher inference efficiency, reaching up to 196.20 img/s on CIFAR10-DVS. These results highlight the efficacy of combining event-driven attention with structured pruning and parameter-efficient position encoding, indicating the potential of SDMFormer for resource-efficient SNN deployment in low-power applications.Keywords
Funding Information
- National Key Research and Development Program of China (2022YFF1202500)
- National Key Research and Development Program of China (2022YFF1202504)
This publication has 54 references indexed in Scilit:
- Directly training temporal Spiking Neural Network with sparse surrogate gradientNeural Networks, 2024
- A gradient descent algorithm for SNN with time-varying weights for reliable multiclass interpretationApplied Soft Computing, 2024
- Error-Aware Conversion from ANN to SNN via Post-training Parameter CalibrationInternational Journal of Computer Vision, 2024
- Conversion of a single-layer ANN to photonic SNN for pattern recognitionScience China Information Sciences, 2023
- Surrogate gradient scaling for directly training spiking neural networksApplied Intelligence, 2023
- Spike Transformer: Monocular Depth Estimation for Spiking Camera2022
- Real Spike: Learning Real-Valued Spikes for Spiking Neural Networks2022
- Multi-label emotion recognition from Indian classical music using gradient descent SNN modelMultimedia Tools and Applications, 2022
- Rethinking the performance comparison between SNNS and ANNSNeural Networks, 2020
- Networks of spiking neurons: The third generation of neural network modelsNeural Networks, 1997