arXiv：cs.LG· Keivan Faghih Niresi, Alice Cicirello, Olga Fink·· 4 小时前AI 评分31
STOIC：基于上下文学习的图能量时间序列不确定性量化残差校准
In-Context Residual Calibration for Uncertainty Quantification of Energy Time Series over Graphs
AI 导读
针对时空图神经网络（STGNN）点预测缺乏可靠不确定性估计的问题，研究者提出 STOIC，一种基于上下文学习的保形启发式事后残差校准框架。STOIC 先用 STGNN 生成点预测，再将时空残差重构为表格表示，借助表格基础模型估计特征条件残差分位数，且无需针对任务重新训练校准模型。该工作已被 TMLR 接收，作者指出 STOIC 本身不提供任意时间依赖下的有限样本无分布覆盖保证。
正文
Abstract:Accurate energy demand forecasting is essential for the reliable operation and planning of modern sustainable energy systems. Spatial-temporal graph neural networks (STGNNs) have recently achieved strong performance in point forecasting by jointly modeling temporal dynamics and relational dependencies across interconnected energy nodes. However, in real-world energy systems, accurate point forecasts alone are insufficient, as operators also require reliable uncertainty estimates to support risk-aware decision-making, grid stability, and operational planning under uncertainty. Conformal prediction provides a principled and model-agnostic framework for uncertainty quantification under exchangeability assumptions, making it particularly attractive for safety-critical energy applications. However, existing conformal prediction approaches often fail to fully capture the complex spatial-temporal structure of energy systems. To address these limitations, we propose STOIC (Spatial-Temporal uncertainty quantificatiOn via In-Context learning), a conformal-inspired post-hoc residual calibration framework that integrates graph-based forecasting with the in-context learning capabilities of tabular foundation models. STOIC first generates point forecasts using an STGNN and subsequently reformulates spatial-temporal residuals into a tabular representation suitable for in-context learning. Leveraging a tabular foundation model, STOIC estimates feature-conditional residual quantiles without task-specific retraining of the calibration model, effectively capturing both sequential and relational dependencies. STOIC should be viewed as a conformal-inspired residual calibration method and does not by itself provide a finite-sample distribution-free coverage guarantee under arbitrary temporal dependence...
| Comments: | Accepted to Transactions on Machine Learning Research (TMLR) |
| Subjects: | Machine Learning (cs.LG); Machine Learning (stat.ML) |
| Cite as: | arXiv:2606.31804 [cs.LG] |
| (or arXiv:2606.31804v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2606.31804 arXiv-issued DOI via DataCite |
Submission history
From: Keivan Faghih Niresi [view email]
[v1]
Tue, 30 Jun 2026 15:22:11 UTC (398 KB)
[v2]
Wed, 7 Oct 2026 09:51:19 UTC (485 KB)
来源：arXiv：cs.LG · arxiv.org