ReCycle：用于长时序列预测的快速高效残差循环Transformer
ReCycle: Fast and Efficient Long Time Series Forecasting with Residual Cyclic Transformers
AI总结：
ReCycle通过主循环压缩降低长时序列注意力计算开销，并结合平滑平均残差学习，在提升预测精度的同时将能耗和运行时间降低一个数量级以上，适用于边缘设备。
摘要：
近年来，Transformer在长时序列预测领域日益受到关注，在多种应用场景中提升了预测精度。然而，在追求更高预测性能的过程中，模型架构的开销变得愈发沉重，导致许多模型的计算需求对于大多数实际应用而言难以承受。为了弥合高复杂度方法与现实计算资源之间的差距，我们提出了残差循环Transformer——ReCycle。ReCycle利用主循环压缩来解决长时序列中注意力机制的计算复杂性问题。通过从精细化的平滑平均技术中学习残差，ReCycle在多种应用场景中超越了当前最先进的预测精度。由简单但稳健的平滑平均技术所保证的可靠且可解释的回退行为，进一步降低了用户接受的障碍。与此同时，我们的方法将运行时间和能耗降低了一个数量级以上，使得训练和推理在低性能、低功耗的边缘计算设备上均可行。代码可在 https://github.com/Helmholtz-AI-Energy/ReCycle 获取。
英文摘要：
Transformers have recently gained prominence in long time series forecasting by elevating accuracies in a variety of use cases. Regrettably, in the race for better predictive performance the overhead of model architectures has grown onerous, leading to models with computational demand infeasible for most practical applications. To bridge the gap between high method complexity and realistic computational resources, we introduce the Residual Cyclic Transformer, ReCycle. ReCycle utilizes primary cycle compression to address the computational complexity of the attention mechanism in long time series. By learning residuals from refined smoothing average techniques, ReCycle surpasses state-of-the-art accuracy in a variety of application use cases. The reliable and explainable fallback behavior ensured by simple, yet robust, smoothing average techniques additionally lowers the barrier for user acceptance. At the same time, our approach reduces the run time and energy consumption by more than an order of magnitude, making both training and inference feasible on low-performance, low-power and edge computing devices. Code is available at https://github.com/Helmholtz-AI-Energy/ReCycle
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。
