休眠脉冲神经元：一种用于高效脉冲神经网络的状态驱动机制
The Dormant Spiking Neuron: A State-Driven Mechanism for Efficient Spiking Neural Networks
AI总结：
提出休眠脉冲神经元（DSN）模型，通过状态驱动动态门实现神经元休眠以减少冗余计算，在多数据集上可降低40%-70%能耗，在ImageNet上还提升基线SNN准确率1.34%。
摘要：
脉冲神经网络（SNN）为高能效人工智能提供了一条极具吸引力的途径，其根本基础是突触前输入驱动的稀疏性。然而，存在一个被长期忽视的巨大计算开销来源：突触后神经元的连续亚阈值动态。定量分析显示，在经过ImageNet训练的ResNet-18模型中，多达44%的神经元处于深度抑制状态，尽管其放电概率可忽略不计，却仍带来持续的计算负担。为解决这一效率问题，本文提出了休眠脉冲神经元（DSN）模型。通过引入状态驱动的动态门，DSN允许神经元在膜电位低于预定义负阈值时进入休眠状态，在实例级别动态修剪冗余计算。关键在于，这种休眠机制被直接整合到神经元动态中，并通过基于代理梯度的随时间反向传播进行端到端训练，使网络能够主动学习鲁棒的休眠策略，自适应地保留关键信息流。在静态数据集和神经形态数据集上开展的大量实验表明，DSN框架可将计算能耗降低40%至70%以上。值得注意的是，在大规模ImageNet数据集上，DSN在可比的报告成本降低幅度下，性能优于最先进的静态修剪技术，同时将对应的基线SNN准确率提升了1.34%。通过建立与传统输入驱动方法正交的可训练、状态驱动稀疏性范式，DSN释放了协同效率提升，为推进神经形态计算提供了有效方法。
英文摘要：
Spiking neural networks (SNNs) offer a compelling pathway toward energy-efficient artificial intelligence, fundamentally anchored in presynaptic input-driven sparsity. However, a substantial yet chronically overlooked source of computational overhead persists: the continuous subthreshold dynamics of postsynaptic neurons. Quantitative analysis reveals that up to 44% of neurons in an ImageNet-trained ResNet-18 model reside in a deeply inhibited state, imposing a persistent computational burden despite having a negligible probability of firing. To address this inefficiency, the Dormant Spiking Neuron (DSN) model is introduced. By incorporating a state-driven dynamic gate, the DSN allows a neuron to enter a dormant state when its membrane potential falls below a predefined negative threshold, dynamically pruning redundant computations at the instance level. Crucially, this dormancy mechanism is directly integrated into the neuronal dynamics and trained end-to-end via backpropagation through time using surrogate gradients, enabling the network to actively learn a robust dormancy policy that adaptively preserves critical information flow. Extensive experiments across static and neuromorphic datasets demonstrate that the DSN framework consistently reduces computational energy by 40% to over 70%. Notably, on the large-scale ImageNet dataset, the DSN outperforms state-of-the-art static pruning techniques at comparable reported cost reductions and simultaneously improves the corresponding baseline SNN accuracy by 1.34%. By establishing a trainable, state-driven sparsity paradigm orthogonal to conventional input-driven methods, the DSN unlocks synergistic efficiency gains, presenting an effective approach for advancing neuromorphic computing.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
