发表机构
Zhejiang Gongshang University; Sussex Artificial Intelligence Institute; School of Information and Electronic Engineering; University of Auckland; School of Computer Science（浙江工商大学; 萨塞克斯人工智能研究所; 信息与电子工程学院; 奥克兰大学; 计算机科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对概率多元时间序列预测中现有扩散方法的信息异质性问题，提出DynG-Diff框架，通过两阶段训练、状态感知策略网络及动态权重设计，实现更优预测性能与鲁棒性。
AI 中文摘要
概率多元时间序列（MTS）预测对复杂动力系统建模至关重要，但现有基于扩散的方法依赖特定任务的条件范式，缺乏灵活性，且难以应对固有的“信息异质性”——即不同变量间噪声水平和演化模式存在显著差异。为解决该问题，本文提出DynG-Diff，一种面向概率多元时间序列预测的变量敏感动态引导扩散框架：（1）DynG-Diff采用两阶段分离训练策略，使用无条件扩散骨干网络对多元时间序列的联合分布进行建模；（2）DynG-Diff引入轻量型状态感知策略网络，从实时噪声状态和单步去噪估计中自适应推断变量可靠性，输出动态引导强度矩阵；（3）DynG-Diff将该动态权重数学表述为观测分布的局部精度，从而在推理阶段对高置信度变量提供精准引导，同时过滤异常噪声的干扰。在真实世界基准上开展的大量实验表明，DynG-Diff的概率预测性能与当前最优的条件扩散模型相比具有竞争力，且在严重观测噪声下的鲁棒性得到提升。该方法的实现代码可在：this https URL 获取。
英文摘要
Probabilistic multivariate time series (MTS) forecasting is crucial for modeling complex dynamical systems. However, existing diffusion-based methods rely on task-specific conditional paradigms that lack flexibility and struggle with inherent "information heterogeneity"--the significantly varying noise levels and evolutionary patterns across variables. To address this, we propose DynG-Diff, a variable-sensitive dynamic guidance diffusion framework for probabilistic multivariate time-series forecasting: (1) DynG-Diff adopts a two-stage separated training strategy and uses an unconditional diffusion backbone to model the joint distribution of multivariate time series. (2) DynG-Diff introduces a lightweight state-aware policy network that adaptively infers variable reliability from real-time noisy states and one-step denoising estimates, outputting a dynamic guidance strength matrix. (3) DynG-Diff mathematically formulates this dynamic weight as the local precision of the observation distribution, enabling precise guidance for high-confidence variables during inference while filtering out interference from anomalous noise. Extensive experiments on real-world benchmarks demonstrate competitive probabilistic forecasting performance against state-of-the-art conditional diffusion models and improved robustness under severe observation corruption.The implementation code is available at: https://github.com/TT-20011031/DynG-Diff
