arXiv 2604.164962026-04-14cs.NEcs.AIcs.LG

无梯度的脉冲神经网络持续学习中的间隔正则化

Gradient-Free Continual Learning in Spiking Neural Networks via Inter-Spike Interval Regularization

University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Indian Institute of Technology Delhi（印度理工学院德里）；University of Texas at El Paso（得克萨斯大学埃尔帕索分校）；National Center for Supercomputing Applications（国家超级计算应用中心）

Samrendra Roy, Kazuma Kobayashi, Souvik Chakraborty, Sajedul Talukder, Syed Bahauddin Alam

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出一种无梯度的脉冲神经网络持续学习方法，通过间隔变化系数来评估突触重要性，避免遗忘并提升实际神经形态芯片上的性能。

摘要：

持续学习，即在不遗忘先前知识的情况下逐步学习新任务的能力，对于部署神经网络在动态真实世界环境中至关重要，从核数字孪生监控到电网边缘故障检测。现有突触重要性方法，如弹性权重巩固（EWC）和突触智能（SI），依赖梯度计算，使其与缺乏反向传播支持的神经形态硬件不兼容。我们提出了ISI-CV，这是首个无梯度的突触重要性度量方法，用于SNN持续学习，源自间隔间隔（ISIs）的变化系数（CV）。那些规律放电（低CV）的神经元编码稳定、任务相关特征，并受到保护不被覆盖；放电不规律的神经元则被允许自由适应。ISI-CV仅需要脉冲时间计数器和整数运算，这些都在每个神经形态芯片中都是原生的。我们在四个难度递增的基准上进行了评估：Split-MNIST、Permuted-MNIST、Split-FashionMNIST和Split-N-MNIST，使用真实的动态视觉传感器（DVS）事件数据。在三个种子下，ISI-CV在Split-MNIST和Split-FashionMNIST上实现了零遗忘（AF = 0.000 ± 0.000），在Permuted-MNIST上接近零遗忘（AF = 0.001 ± 0.000），并在真实的神经形态DVS数据上实现了最高准确率和最低遗忘（AA = 0.820 ± 0.012，AF = 0.221 ± 0.014）。在N-MNIST上，基于梯度的方法产生不可靠的重要性估计，并且表现比无正则化更差；ISI-CV通过设计避免了这种失败。

英文摘要：

Continual learning, the ability to acquire new tasks sequentially without forgetting prior knowledge, is essential for deploying neural networks in dynamic real-world environments, from nuclear digital twin monitoring to grid-edge fault detection. Existing synaptic importance methods, such as Elastic Weight Consolidation (EWC) and Synaptic Intelligence (SI), rely on gradient computation, making them incompatible with neuromorphic hardware that lacks backpropagation support. We propose ISI-CV, the first gradient-free synaptic importance metric for SNN continual learning, derived from the Coefficient of Variation (CV) of Inter-Spike Intervals (ISIs). Neurons that fire regularly (low CV) encode stable, task-relevant features and are protected from overwriting; neurons with irregular firing are permitted to adapt freely. ISI-CV requires only spike time counters and integer arithmetic, all of which are native to every neuromorphic chip. We evaluate on four benchmarks of increasing difficulty: Split-MNIST, Permuted-MNIST, Split-FashionMNIST, and Split-N-MNIST using real Dynamic Vision Sensor (DVS) event data. Across three seeds, ISI-CV achieves zero forgetting (AF = 0.000 +/- 0.000) on Split-MNIST and Split-FashionMNIST, near-zero forgetting on Permuted-MNIST (AF = 0.001 +/- 0.000), and the highest accuracy with the lowest forgetting on real neuromorphic DVS data (AA = 0.820 +/- 0.012, AF = 0.221 +/- 0.014). On N-MNIST, gradient-based methods produce unreliable importance estimates and perform worse than no regularization; ISI-CV avoids this failure by design.

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。