AI 中文总结
DiffPTS通过重新推导位置-尺度噪声模型下的ELBO，提出统一联合训练目标，实现端到端优化，在概率时间序列预测中平均CRPS/MSE降低超14.53%/16.55%。
AI 中文摘要
概率时间序列预测需要建模和预测复杂且时变的分布。近期，基于去噪扩散概率模型（DDPM）的方法通过为扩散过程配备预训练的均值和方差估计器以适应分布偏移，显示出良好的前景。然而，这些方法通常遵循标准的DDPM框架，仅考虑证据下界（ELBO）的部分组件，将估计器的训练视为与变分推断框架分离的既定回归任务。为解决此问题，我们在位置-尺度噪声模型（LSNM）下重新思考ELBO，发现其自然地为估计器引出一个高斯负对数似然目标，并固有地定义了一个联合训练目标，统一了近期用于概率预测的扩散范式。基于这一原则性的ELBO重构，我们提出DiffPTS，一个通用框架，能够对ELBO内的所有组件进行端到端优化。在多个基准上，DiffPTS持续优于近期模型，与现有基于扩散的方法相比，平均CRPS/MSE降低超过14.53%/16.55%，实现了最先进的性能。代码可在该https URL获取。
英文摘要
Probabilistic time series forecasting requires modeling and predicting complex and time-varying distributions. Recently, Denoising Diffusion Probabilistic Model (DDPM)-based approaches have shown promise by equipping the dif- fusion process with pretrained mean and variance estimators to accommodate distributional shift. However, these methods typically follow the standard DDPM framework and consider only partial components of the evidence lower bound (ELBO), treating the training of estimators as designed regression tasks separate from the variational inference framework. To address this, we rethink the ELBO under the Location-Scale Noise Model (LSNM) and find that it naturally induces a Gaussian negative log likelihood objective for the estimators and inherently defines a joint training objective that unifies recent diffusion paradigms for probabilistic forecasting. Building on this principled ELBO reformulation, we propose Diff- PTS, a general framework that enables end-to-end optimization of all components within the ELBO. Across multiple benchmarks, DiffPTS consistently outperforms recent models, achieving state-of-the-art performance with an average CRPS/MSE reduction of over 14.53%/16.55% compared to existing diffusion-based methods. The code is available at https://github.com/wwy155/DiffPTS.
CommentsAccepted as NeurIPS 2026 Poster
