SNN转换连续控制时性能下降，因误差随时间累积放大。
Error Amplification Limits ANN-to-SNN Conversion in Continuous Control
- 用跨步残差电位初始化抑制动作误差的时序相关性。
- 在向量与视觉观测任务中恢复了转换后的性能损失。
- 适合做强化学习中SNN部署的开发者参考。
脉冲神经网络（SNN）可通过转换已训练好的人工神经网络（ANN）实现良好性能，避免额外昂贵训练。这一特性在强化学习（RL）中尤为吸引人，因环境交互训练成本高且存在安全隐患。然而现有转换方法在连续控制任务中表现不佳，且缺乏合适基线。我们识别出误差放大是根本原因：微小的动作近似误差在决策步骤间产生时序相关性，引发状态分布累积偏移，导致性能严重退化。为此，我们提出无梯度的跨步残差电位初始化（CRPI），通过跨步骤传递残差膜电位抑制时序相关误差。在含向量与视觉观测的连续控制基准测试中，CRPI可无缝集成至现有转换流程，并显著恢复性能损失。结果表明连续控制是ANN-to-SNN转换的关键挑战场景，微小误差可能被强烈放大并影响最终表现。代码已公开于https://github.com/xuzijie32/ANN2SNN-CRPI。
原文摘要 · Abstract (English)
Spiking Neural Networks (SNNs) can achieve competitive performance by converting already existing well-trained Artificial Neural Networks (ANNs), avoiding further costly training. This property is particularly attractive in Reinforcement Learning (RL), where training through environment interaction is expensive and potentially unsafe. However, existing conversion methods perform poorly in continuous control, where suitable baselines are largely absent. We identify error amplification as the key cause: small action approximation errors become temporally correlated across decision steps, inducing cumulative state distribution shift and severe performance degradation. To address this issue, we propose Cross-Step Residual Potential Initialization (CRPI), a lightweight gradient-free mechanism that carries over residual membrane potentials across decision steps to suppress temporally correlated errors. Experiments on continuous control benchmarks with both vector and visual observations demonstrate that CRPI can be integrated into existing conversion pipelines and substantially recovers lost performance. Our results highlight continuous control as a critical and challenging benchmark for ANN-to-SNN conversion, where small errors can be strongly amplified and impact performance. Code is available at https://github.com/xuzijie32/ANN2SNN-CRPI.
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。