SpikySpace: 一种用于节能时间序列预测的脉冲状态空间模型
SpikySpace: A Spiking State Space Model for Energy-Efficient Time Series Forecasting
AI总结：
SpikySpace通过脉冲选择扫描和高效近似方法，实现了在能耗降低96.1%的同时提升时间序列预测准确性3.0%的高效节能模型。
摘要：
在交通管理、工业监控等领域中，时间序列预测往往需要在资源有限的边缘设备上进行实时、节能的处理。脉冲神经网络（SNNs）提供事件驱动计算和超低功耗，并已被提出用于此领域。然而，现有基于SNN的时间序列预测器往往使用复杂的Transformer块。为了解决这一问题，我们提出了SpikySpace，一种脉冲状态空间模型（SSM），通过脉冲选择扫描将注意力块中的二次成本降低到线性时间。进一步，我们引入了PTsoftplus和PTSiLU，两种高效的SiLU和Softplus近似方法，用简单的位移操作代替了昂贵的指数和除法操作。在四个多变量时间序列基准上评估，SpikySpace在准确性上比领先的SNN高出多达3.0%，同时将能耗降低超过96.1%。作为首个完全脉冲的状态空间模型，SpikySpace将神经形态效率与现代序列建模相结合，为高效时间序列预测系统开辟了实际路径。我们的代码可在https://anonymous.4open.science/r/SpikySpace上获得。
英文摘要：
Time-series forecasting in domains like traffic management and industrial monitoring often requires real-time, energy-efficient processing on edge devices with limited resources. Spiking neural networks (SNNs) offer event-driven computation and ultra-low power and have been proposed for use in this space. Unfortunately, existing SNN-based time-series forecasters often use complex transformer blocks. To address this issue, we propose SpikySpace, a spiking state-space model (SSM) that reduces the quadratic cost in the attention block to linear time via spiking selective scanning. Further, we introduce PTsoftplus and PTSiLU, two efficient approximations of SiLU and Softplus that replace costly exponential and division operations with simple bit-shifts. Evaluated on four multivariate time-series benchmarks, SpikySpace outperforms the leading SNN in terms of accuracy by up to 3.0% while reducing energy consumption by over 96.1%. As the first fully spiking state-space model, SpikySpace bridges neuromorphic efficiency with modern sequence modeling, opening a practical path toward efficient time series forecasting systems. Our code is available at https://anonymous.4open.science/r/SpikySpace.
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。
