用随机采样提升长期预测精度，降低显存占用。
Improving Long-term Autoregressive Spatiotemporal Predictions: A Proof of Concept with Fluid Dynamics
- 引入随机推进框架，结合真实数据与预测数据训练
- 长期预测误差降低，显存使用稳定不增长
- 适合资源受限的复杂系统模拟场景
数据驱动方法正成为传统数值预报的高效替代方案，推理快且计算成本低。然而，对于复杂系统，长期预测常因误差累积而性能下降；自回归训练虽有效，但需大量显存，且可能牺牲短期表现。本文提出随机推进（SPF）框架，在保留一步预测训练的同时实现多步学习。SPF通过随机策略从模型预测中构建补充数据集，并与真实数据融合，平衡短时与长时性能，同时减少过拟合。多步预测在每个训练周期间预计算完成，避免存储完整展开序列，保持内存稳定。在Burgers方程和浅水模型基准测试中，SPF相比传统自回归方法显著提升长期预测精度，同时降低显存需求，展现出在资源受限和复杂模拟中的应用潜力。
原文摘要 · Abstract (English)
Data-driven methods are emerging as efficient alternatives to traditional numerical forecasting, offering fast inference and lower computational cost. Yet, for complex systems, long-term accuracy often deteriorates due to error accumulation, and autoregressive training (though effective) demands large GPU memory and may sacrifice short-term performance. We propose the Stochastic PushForward (SPF) framework, which retains one-step-ahead training while enabling multi-step learning. SPF builds a supplementary dataset from model predictions and combines it with ground truth via a stochastic acquisition strategy, balancing short- and long-term performance while reducing overfitting. Multi-step predictions are precomputed between epochs, keeping memory usage stable without storing full unrolled sequences. Experiments on the Burgers' equation and the Shallow Water benchmark show that SPF achieves higher long-term accuracy than autoregressive methods while lowering memory requirements, making it promising for resource-limited and complex simulations.
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。