探索循环脉冲神经网络的打盹范式
Exploring napping paradigm for Recurrent Spiking Neural Networks
浏览论文内容
中文总结 AI 辅助
本文提出受生物启发的“打盹”微睡眠范式（比例权重缩放加随机膜噪声），在无监督循环SNN上验证其能匹配权重归一化的精度，同时降低模型复杂度并增强表示结构。
中文摘要 AI 辅助
生物有机体通过平衡其内部世界模型上的两个相互竞争的需求来最小化自由能：该模型必须足够准确以预测感觉输入，但又必须足够简单以泛化到输入之外。两种机制在离线状态下调节这种平衡：睡眠通过逐渐的突触缩减来降低复杂度，而随机噪声则削弱精度，从而放松感觉输入对突触重组所施加的约束。工程化的脉冲神经网络（SNNs）未处理这种平衡，而是倾向于瞬时、无噪声的权重归一化。本文研究了一个假设：一种受生物学启发的微睡眠范式，即打盹——结合比例权重缩放与连续随机膜活动——能够在降低模型复杂度的同时复制归一化的稳定性。我们在一个通过基于迹的脉冲时序依赖可塑性（STDP）在Gabor预处理的MNIST上训练的无监督循环SNN中评估了这一点。我们通过扫描打盹的持续时间和膜噪声水平，在三种正则化机制下调整打盹，然后将最佳配置与权重归一化进行比较。在所有三种机制中，调整良好的打盹匹配了归一化的准确性：准确性在短暂持续时间和低噪声时达到峰值，然后随着任一者的增加而单调下降。聚类则出现分歧，最强的几何分离出现在较长的持续时间和较高的噪声下——自由能的两个项相互分离，准确性奖励数据拟合，而结构奖励逐渐、嘈杂缩减所诱导的更简单表示。这种收益带来了归一化所避免的模拟成本，因此打盹在表示结构而非原始分类效率是优先考虑的情况下最具吸引力。
英文摘要
Biological organisms minimize free energy by balancing two competing demands on their internal world model: it must be accurate enough to predict sensory input, yet simple enough to generalize beyond it. Two mechanisms regulate this balance offline: sleep reduces complexity through gradual synaptic downscaling, while stochastic noise attenuates precision, relaxing the constraint sensory input imposes on synaptic reorganization. Engineered Spiking Neural Networks (SNNs) leave this balance unaddressed, favoring instantaneous, noiseless weight normalization instead. This paper investigates the hypothesis that a biologically inspired micro-sleep paradigm, napping -- combining proportional weight scaling with continuous stochastic membrane activity -- can replicate the stability of normalization while shedding model complexity. We evaluate this in an unsupervised recurrent SNN trained via trace-based spike-timing-dependent plasticity (STDP) on Gabor-preprocessed MNIST. We tune napping across three regularization regimes by sweeping its duration and membrane noise level, then compare the best configuration against weight normalization. Across all three regimes, well-tuned napping matches the accuracy of normalization: accuracy peaks at brief durations and low noise, then declines monotonically as either grows. Clustering diverges, with the strongest geometric separation arising at longer durations and higher noise -- the two terms of free energy pulling apart, accuracy rewarding data fit and structure rewarding the simpler representation that gradual, noisy downscaling induces. This gain carries a simulation cost normalization avoids, so napping is most compelling where representational structure, rather than raw classification efficiency, is the priority.
发表机构
- The Norwegian University of Life Sciences（挪威生命科学大学）
- Østfold University of Applied Sciences（东福尔应用科学大学）
机构由 AI 辅助整理，请以论文原文为准。
