用时间扭曲方法迁移RNN模型，高效适配不同变化速度的物理系统。
Time-Warping Recurrent Neural Networks for Transfer Learning
- 通过时间缩放重构RNN时间维度，实现跨时间尺度的模型迁移。
- 在1小时、100小时、1000小时时间尺度下预测燃料含水率，精度接近主流方法。
- 仅调整少量参数即达成良好性能，适合数据稀缺场景的模型复用。
动力系统描述物理系统随时间演化的过程。环境差异会导致物理过程加速或减速。本文提出一种基于时间扭曲（time-warping）的RNN迁移学习新方法，通过重标定时间尺度来适应不同演化速度。研究证明，对于一类称为时滞模型的一阶线性微分方程，LSTM可任意逼近其动态行为，且经时间扭曲后仍保持近似精度。该方法在野火建模中的燃料含水率（FMC）预测任务中验证：先在特征时间尺度为10小时的丰富数据上预训练LSTM模型，再通过时间扭曲迁移至1小时、100小时和1000小时特征时间尺度的燃料。与多种现有迁移学习方法对比，时间扭曲法仅修改极小部分参数，即可达到相近预测精度。
原文摘要 · Abstract (English)
Dynamical systems describe how a physical system evolves over time. Physical processes can evolve faster or slower in different environmental conditions. We use time-warping as rescaling the time in a model of a physical system. This thesis proposes a new method of transfer learning for Recurrent Neural Networks (RNNs) based on time-warping. We prove that for a class of linear, first-order differential equations known as time lag models, an LSTM can approximate these systems with any desired accuracy, and the model can be time-warped while maintaining the approximation accuracy. The Time-Warping method of transfer learning is then evaluated in an applied problem on predicting fuel moisture content (FMC), an important concept in wildfire modeling. An RNN with LSTM recurrent layers is pretrained on fuels with a characteristic time scale of 10 hours, where there are large quantities of data available for training. The RNN is then modified with transfer learning to generate predictions for fuels with characteristic time scales of 1 hour, 100 hours, and 1000 hours. The Time-Warping method is evaluated against several known methods of transfer learning. The Time-Warping method produces predictions with an accuracy level comparable to the established methods, despite modifying only a small fraction of the parameters that the other methods modify.
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。