用扩散模型替代传统高斯过渡，提升时序建模的拟合与预测能力。
Diffusion-Driven State Space Models
- 用扩散模型替代高斯分布建模潜在状态转移
- 在多模态时序数据上表现优于当前最优深度状态空间模型
- 适用于需精准捕捉复杂动态的时序预测任务
在诸多领域，研究者希望模型既能准确预测，又能忠实刻画潜在系统动态。现有方法通常在两者间妥协：深度状态空间模型常假设高斯潜变量转移，限制了拟合与预测能力；而扩散模型虽表达能力强，却缺乏对潜在动态的合理推断。为此，本文提出扩散驱动的状态空间模型（DDSSM），将传统高斯转移分布替换为扩散模型。该模型解决了如何在序列数据上联合训练自编码器与扩散模型的开放问题，扩展了时序潜在扩散模型的研究。实验表明，DDSSM在模拟的具有多模态转移的时序数据上，拟合与预测性能均优于当前最先进的深度状态空间模型。
原文摘要 · Abstract (English)
In many domains, practitioners seek models that produce accurate forecasts while faithfully capturing latent system dynamics. Existing approaches typically sacrifice one of these goals: deep state space models often assume Gaussian latent transitions, limiting fit and forecasting, while diffusion models are highly expressive but lack principled inference for the underlying dynamics. To combine the strengths of both, we introduce the Diffusion-Driven State Space Model (DDSSM), which replaces the conventional Gaussian transition distribution with a diffusion model. Our DDSSM resolves the open problem of how to jointly train an autoencoder and a diffusion model on sequential data, thereby extending the literature on latent diffusion models for time series. Moreover, we find that the DDSSM empirically outperforms a state-of-the-art deep SSM at fitting and forecasting a simulated time series with multimodal transitions.
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。