发表机构
College of Computer and Information Technology, China Three Gorges University; City University of New York; University of California, Berkeley; Emory University（三峡大学计算机与信息学院; 纽约城市大学; 加州大学伯克利分校; 埃默里大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CoRe提出一种模型无关的损失函数，通过频率相干性和低秩关系图约束替代逐点误差，提升多元时间序列直接预测的结构保真度，实验证明其能稳定增强多种基线模型。
AI 中文摘要
直接预测已成为多元时间序列预测的标准范式，因为它能在单次前向传播中预测完整的未来时域。然而，其训练目标通常仍被分解为诸如均方误差（MSE）之类的逐点误差。此类目标提供了稳定的监督信号，但并未显式保留未来轨迹的结构：每个变量内部的时间相干性以及跨变量的关系一致性都可能被削弱。我们提出CoRe，一种用于直接多元预测的模型无关学习目标。CoRe用两个输出空间约束取代逐点监督：一个频率相干性损失，用于对齐预测谱与目标谱；以及一个低秩关系图损失，用于在目标导出的PCA子空间中匹配采样的成对差异。由此产生的目标不引入可训练参数，且只需更改损失函数即可应用于现有的预测骨干网络。在标准基准上的实验表明，CoRe提升了强基线的性能，与近期的预测目标相比具有竞争力，并在不同骨干网络、数据集和超参数设置下整体保持一致的有效性。
英文摘要
Direct forecasting has become a standard paradigm for multivariate time-series forecasting because it predicts the full future horizon in a single pass. However, its training objective is often still decomposed into pointwise errors such as MSE. Such objectives provide stable supervision, but they do not explicitly preserve the structure of the future trajectory: temporal coherence within each variable and relational consistency across variables can both be weakened. We propose CoRe, a model-agnostic learning objective for direct multivariate forecasting. CoRe replaces pointwise supervision with two output-space constraints: a frequency coherence loss that aligns predicted and target spectra, and a low-rank relational graph loss that matches sampled pairwise differences in a target-derived PCA subspace. The resulting objective introduces no trainable parameters and can be applied to existing forecasting backbones by changing only the loss. Experiments on standard benchmarks show that CoRe improves strong baselines, compares favorably with recent forecasting objectives, and remains effective across different backbones, datasets, and hyperparameter settings overall consistently.
CommentsAccepted at the International Conference on Neural Information Processing (ICONIP 2026)
