arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.21776cs.LG
收藏

用于合成时间序列分类数据集的因果DAG先验

A Causal DAG Prior for Synthetic Time-Series Classification Datasets

Massachusetts Institute of Technology, Cambridge, MA, USA(麻省理工学院)

Franco Martino O'Rourke, Ana Trisovic, Dimitris Bertsimas

AI总结:

提出一种因果DAG先验,从随机采样的有向无环图生成多变量、多类别的时间序列分类数据集,包含跨模态因果结构,微调TabPFN v2.5后在UCR/UEA数据集上显著优于基线。

摘要:

先验数据拟合网络学习由其训练先验诱导的后验预测;将这一范式推广到多变量时间序列分类因此需要一个能够生成具有时间结构的完整标记数据集的合成生成器。我们引入一种因果先验,该先验从随机采样的有向无环图(DAG)中合成每个数据集,该DAG覆盖两种模态(表格属性和时间序列)上的类型化节点,原生生成具有跨通道、时间步和标签的跨模态因果结构的多变量、多类别TSC数据集,这是现有合成先验未涉及的机制。为验证该先验,我们在最小化调整的情况下微调TabPFN v2.5,并在TabPFN的运行范围内对75个UCR/UEA数据集进行评估。在我们的生成器上微调显著优于未修改的上游模型以及相同先验的纯表格消融(ROC-AUC上Wilcoxon符号秩检验$p=3.0\ imes 10^{-8}$),从而分离出跨模态时间结构的贡献。

英文摘要:

A Prior-data fitted Network learns the posterior predictive induced by its training prior; bringing this paradigm to multivariate time-series classification therefore calls for a synthetic generator that produces complete labelled datasets with temporal structure. We introduce a causal prior that synthesizes each dataset from a randomly sampled DAG over typed nodes across two modalities (tabular attributes and time series), natively producing multivariate, multi-class TSC datasets with cross-modal causal structure across channels, timesteps and labels, a regime not addressed by existing synthetic priors. To validate the prior, we finetune TabPFN v2.5 with minimal adaptations and evaluate on 75 UCR/UEA datasets within TabPFN's operating regime. Finetuning on our generator significantly outperforms both the unmodified upstream model and a tabular-only ablation of the same prior (Wilcoxon signed-rank $p=3.0\times 10^{-8}$ on ROC-AUC), isolating the contribution of the cross-modal temporal structure.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑