arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.23074cs.AIcs.LG

事件签名迁移:从历史事件进行模型无关的预测情景构建

Event Signature Transfer: Model-Agnostic Forecast Scenario Construction from Historical Events

Karthik Sridhar, Aaditya Jain, Murari Mandal, Saurabh Deshpande

首次发表
浏览论文内容

中文总结 AI 辅助

提出事件签名迁移(EST),一种无需训练、模型无关的算子,通过移除趋势和季节性并缩放重定时,将历史事件转化为预测情景,在多个模型上显著降低 WQL。

中文摘要 AI 辅助

预测者通常知道某个事件即将发生,但不知道其影响的形式、规模或时间。我们引入了事件签名迁移(EST),这是一种无需训练、模型无关的算子,可将已完成的过去事件转化为明确的预测情景。EST 移除源事件自身的趋势和季节性,然后将剩余的事件签名进行缩放并重新定时,应用到原生预测上,保留预测的关联结构,并在强度为零时精确还原为原始预测。由于 EST 仅读取输出分位数,因此它适用于任何分位数预测器,在迁移时无需训练,也无需访问模型内部。在 Chronos-2、TimesFM-2.5 和 Toto-2.0 上的十二个真实事件和十个合成情景中,手动配置的 EST 在样本内将真实事件 WQL 降低了 21.7% 至 90%。在 Chronos-2 上,与协变量条件化、激活编辑和原始回放相比,它在十二个匹配比较中领先了十一个。该算子构建情景,但不估计其可能性。

英文摘要

Forecasters often know an event is imminent but not the shape, size, or timing of its effect. We introduce Event Signature Transfer (EST), a training-free, model-agnostic operator that turns a completed past event into an explicit forecast scenario. EST removes a source event's own trend and seasonality, then scales and retimes the remaining event signature onto a native forecast, preserving the forecast's linked structure and reducing to it exactly at zero strength. Because it reads only output quantiles, EST applies to any quantile forecaster, with no training, no model internals, at transfer time. Across twelve real episodes and ten synthetic scenarios on Chronos-2, TimesFM-2.5 and Toto-2.0, manually configured EST reduces real-episode WQL by 21.7-90\% in-sample. On Chronos-2, it leads eleven of twelve matched comparisons against covariate conditioning, activation editing and raw replay. The operator builds a scenario; it does not estimate its likelihood.

发表机构

  • Birla AI Labs(Birla AI 实验室)
  • KIIT Bhubaneswar(KIIT 布巴内斯瓦尔)

机构由 AI 辅助整理,请以论文原文为准。

↑