用未来数据反馈优化预测模型，提升长期依赖捕捉能力。
A Predictive Approach To Enhance Time-Series Forecasting
- 通过未来数据检测与当前预测对比，动态调整模型参数
- 癫痫预测AUC-ROC提升44.8%，非线性系统预测MSE降低23.4%
- 适合需要应对数据分布变化的时序预测场景
准确的时序预测在多个科学和工业领域至关重要，但深度学习模型常难以捕捉长期依赖关系并适应随时间变化的数据分布。本文提出未来引导学习（Future-Guided Learning），一种受预测编码启发的动态反馈机制，用于增强时序事件预测。该方法包含两个模型：一个检测模型分析未来数据以识别关键事件，另一个预测模型基于当前数据进行预测。当两模型输出存在差异时，对预测模型施加更大更新，从而减少意外，实现参数的动态调整。我们在多种任务上验证该方法，使用脑电图（EEG）数据进行癫痫预测时，AUC-ROC提升44.8%；在非线性动力系统预测中（剔除异常值后），均方误差（MSE）下降23.4%。通过引入预测反馈机制，未来引导学习推动了深度学习在时序预测中的应用边界。
原文摘要 · Abstract (English)
Accurate time-series forecasting is crucial in various scientific and industrial domains, yet deep learning models often struggle to capture long-term dependencies and adapt to data distribution shifts over time. We introduce Future-Guided Learning, an approach that enhances time-series event forecasting through a dynamic feedback mechanism inspired by predictive coding. Our method involves two models: a detection model that analyzes future data to identify critical events and a forecasting model that predicts these events based on current data. When discrepancies occur between the forecasting and detection models, a more significant update is applied to the forecasting model, effectively minimizing surprise, allowing the forecasting model to dynamically adjust its parameters. We validate our approach on a variety of tasks, demonstrating a 44.8% increase in AUC-ROC for seizure prediction using EEG data, and a 23.4% reduction in MSE for forecasting in nonlinear dynamical systems (outlier excluded).By incorporating a predictive feedback mechanism, Future-Guided Learning advances how deep learning is applied to time-series forecasting.
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。