提出并行可逆脉冲神经网络，训练推理更快且省内存。
ParaRevSNN: A Parallel Reversible Spiking Neural Network for Efficient Training and Inference
- 通过解耦可逆模块间的依赖，实现块间并行计算。
- 在多个数据集上准确率相当或更高，训练提速35.2%。
- 适合内存受限设备部署，兼顾效率与性能。
可逆脉冲神经网络（RevSNN）通过反向传播时重构前向激活实现内存高效训练，但因严格串行计算导致延迟高。为克服此问题，本文提出并行可逆脉冲神经网络（ParaRevSNN），在保持可逆性的前提下解耦可逆模块间的序列依赖，实现模块间并行。该设计显著加速训练与推理，同时保留可逆性带来的内存优势。在CIFAR10、CIFAR100、CIFAR10-DVS和DVS128 Gesture数据集上的实验表明，ParaRevSNN的准确率与标准RevSNN相当或更优，训练时间最多减少35.2%，推理时间降至18.15%，适用于资源受限场景部署。
原文摘要 · Abstract (English)
Reversible Spiking Neural Networks (RevSNNs) enable memory-efficient training by reconstructing forward activations during backpropagation, but suffer from high latency due to strictly sequential computation. To overcome this limitation, we propose ParaRevSNN, a parallel reversible SNN architecture that decouples sequential dependencies between reversible blocks while preserving reversibility. This design enables inter-block parallelism, significantly accelerating training and inference while retaining the memory-saving benefits of reversibility. Experiments on CIFAR10, CIFAR100, CIFAR10-DVS, and DVS128 Gesture demonstrate that ParaRevSNN matches or exceeds the accuracy of standard RevSNNs, while reducing training time by up to 35.2\% and inference time to 18.15\%, making it well-suited for deployment in resource-constrained scenarios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。