用数据中的重复模式引导扩散模型，提升时间序列预测准确性。
Pattern-Guided Diffusion Models
- 通过典型分析提取数据中的重复模式，并预估下一时刻模式
- 在视觉场和动作捕捉任务上，预测误差降低最高达56.26%
- 动态调整引导强度，适合有周期性结构的时间序列场景
扩散模型在多变量时间序列预测中表现优异，但现有方法较少考虑数据中反复出现的模式。本文提出模式引导扩散模型（PGDM），利用时间数据中的固有模式进行未来预测。PGDM首先通过典型分析提取模式，并估计序列中最可能的下一个模式。通过该模式估计引导预测，使结果更符合已知模式，更具现实性。我们还提出一种基于典型分析的新不确定性量化方法，并根据模式估计的不确定性动态调整引导强度。在视觉场测量和动作捕捉帧预测两个应用中，实验表明，模式引导使PGDM的性能提升最高达40.67%（MAE）/56.26%（CRPS），分别优于基线模型14.12%/14.10%与65.58%/84.83%、93.64%/92.55%。
原文摘要 · Abstract (English)
Diffusion models have shown promise in forecasting future data from multivariate time series. However, few existing methods account for recurring structures, or patterns, that appear within the data. We present Pattern-Guided Diffusion Models (PGDM), which leverage inherent patterns within temporal data for forecasting future time steps. PGDM first extracts patterns using archetypal analysis and estimates the most likely next pattern in the sequence. By guiding predictions with this pattern estimate, PGDM makes more realistic predictions that fit within the set of known patterns. We additionally introduce a novel uncertainty quantification technique based on archetypal analysis, and we dynamically scale the guidance level based on the pattern estimate uncertainty. We apply our method to two well-motivated forecasting applications, predicting visual field measurements and motion capture frames. On both, we show that pattern guidance improves PGDM's performance (MAE / CRPS) by up to 40.67% / 56.26% and 14.12% / 14.10%, respectively. PGDM also outperforms baselines by up to 65.58% / 84.83% and 93.64% / 92.55%.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。