发表机构
University of California San Diego; University of Alberta; University of Alabama（加州大学圣地亚哥分校; 阿尔伯塔大学; 阿拉巴马大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出可分解脉冲神经网络（D-SNN），采用生物启发模块化架构，在减少参数与突触操作的同时，实现鲁棒性与决策透明性，为边缘神经形态计算提供高效方案。
AI 中文摘要
生物神经系统通过分区架构实现了高效率与鲁棒性，相比之下，现代人工神经网络依赖全局纠缠结构，这会模糊决策逻辑并遭受灾难性遗忘。本文提出了一种可分解脉冲神经网络（D-SNN），该模型通过将分类通路结构上隔离为独立专家，消除了全局突触纠缠。经生物启发的推拉损失函数优化后，D-SNN在MNIST、Fashion-MNIST及CIFAR-10/100基准测试中取得了有竞争力的准确率。这种模块化方法的性能与全密集网络相当，但参数数量减少了一个数量级以上；此外，我们的网络工作时的放电率和突触操作次数可低至几个数量级。更重要的是，物理切断专家间的连接为顺序学习期间的灾难性遗忘提供了固有保护，且这些隔离通路会生成可审计的神经信号，提升了决策透明度。这种仿生且可验证的架构为在资源受限的边缘环境中部署确定性神经形态智能奠定了高效基础。
英文摘要
Biological neural systems achieve high efficiency and robustness through compartmentalized architectures. In contrast, modern artificial neural networks rely on globally entangled structures, which obscure decision logic and suffer from catastrophic forgetting. Here, we report a Decomposable Spiking Neural Network (D-SNN) that eliminates global synaptic entanglement by structurally isolating classification pathways into independent experts. Optimized via a bio-inspired push-pull loss function, the D-SNN achieves competitive accuracies on MNIST, Fashion-MNIST, and CIFAR-10/100 benchmarks. This modular approach matches the performance of fully dense networks while utilizing an order of magnitude fewer parameters. In addition, our networks operate with up to several orders of magnitude lower firing rates and fewer synaptic operations. Furthermore, physically severing connections between experts provides inherent protection against catastrophic forgetting during sequential learning. Crucially, these isolated pathways generate auditable neural signals, increasing decision transparency. This biomimetic, verifiable architecture establishes an efficient foundation for deploying deterministic neuromorphic intelligence in resource-constrained edge environments.
Comments39 pages, 7 figures
