发表机构
Kwangwoon University(光云大学)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
该研究针对时间序列预测的历史检索问题,提出基于未来监督的预测相关性重排序方法,在多基准任务中优于SARAF等检索规则,揭示历史相关性具结构化和领域依赖性。
AI 中文摘要
时间序列预测的历史检索通常将过去的相似性作为实用性的替代指标。本文提出了一个不同的问题:对于一个查询,哪些历史示例应被认为是重要的?我们将预测相关性定义为在推理时信息条件下的预期未来效用,仅在训练期间使用已实现的未来作为特权监督。归一化模式检索器首先形成粗略候选集,轻量级残差多层感知机(MLP)学习列表式未来兼容性目标,同时保持推理时评分严格仅使用过去数据。我们的方法保留基于相似性的候选生成,但通过更具预测性的相关性标准对候选进行重新排序。最优相关性可分解为候选级效用和查询特定兼容性,这催生了候选先验和打乱未来控制。在六个基准测试中,重新排序器改进了模式检索,同时揭示了候选全局、查询特定及混合相关性机制。在所有12项验证任务中,它改进了模式检索并优于匹配协议的平稳性感知检索增强时间序列预测(SARAF)检索规则。架构匹配的 ablation 实验表明,正确的未来监督而非单独的 MLP 或添加的上下文,驱动了查询特定机制下的性能提升。替代相似性实验显示,强的最后值锚定 L2 规则在某些领域仍更优,而未来监督相关性在诊断显示存在查询特定相关性的领域(尤其是 Solar 数据集)表现尤为突出。候选池诊断显示,这种对比并非仅由粗略模式检索解释。总体而言,历史相关性是结构化且依赖领域的,而非受普遍优越的检索规则支配。
英文摘要
Retrieval-augmented time-series forecasting typically selects historical examples by similarity between observed pasts, although similar pasts can evolve differently. We propose Predictive Relevance Retrieval (PRR), which uses realized future compatibility as privileged supervision to learn a retrieval function that remains strictly past-only at inference. PRR combines Pearson retrieval with a futuresupervised predictive representation to expand candidate support, then reranks the union using statistical and learned pair relations. Across six datasets and four long horizons, PRR improves Pearson retrieval in 23 of 24 conditions, reducing AnalogFutureMSE by 26.7% on average. A candidate-budget-matched variant, PRR-B100, retains nearly the same retrieval improvement while using at most 100 candidates at inference, showing that the gain is not explained simply by a larger candidate pool. We then connect the retriever to five frozen forecasting backbones using validation-calibrated trust. Downstream effects are heterogeneous, and calibration primarily reduces harmful retrieval use rather than making improved retrieval universally beneficial. These results show that predictive relevance and forecast utility are empirically distinct objectives.