改进时间序列建模框架，通过显式噪声建模提升预测与补全性能。
Alternators With Noise Models
- 引入噪声模型显式建模潜在与观测轨迹的噪声，增强灵活性。
- 在密度估计、缺失值填补和预测任务上优于Mamba、ScoreGrad等基线。
- 适合需要高精度时间序列建模的研究者与工业应用开发者。
Alternator 近期被提出作为建模时变数据的框架，已在复杂时间序列任务中表现优于状态空间模型和扩散模型。本文提出新模型 Alternator++，通过借鉴扩散模型中的噪声建模思想，显式建模采样潜在与观测轨迹所用的噪声项，从而提升传统 Alternator 的灵活性。Alternator++ 优化 Alternator 损失与噪声匹配损失之和，后者强制两个噪声模型生成的噪声轨迹逼近真实生成观测与潜在轨迹的噪声。实验表明，该方法在密度估计、时间序列插补与预测任务中均显著优于 Mambas、ScoreGrad 与 Dyffusion 等强基线。
原文摘要 · Abstract (English)
Alternators have recently been introduced as a framework for modeling time-dependent data. They often outperform other popular frameworks, such as state-space models and diffusion models, on challenging time-series tasks. This paper introduces a new Alternator model, called Alternator++, which enhances the flexibility of traditional Alternators by explicitly modeling the noise terms used to sample the latent and observed trajectories, drawing on the idea of noise models from the diffusion modeling literature. Alternator++ optimizes the sum of the Alternator loss and a noise-matching loss. The latter forces the noise trajectories generated by the two noise models to approximate the noise trajectories that produce the observed and latent trajectories. We demonstrate the effectiveness of Alternator++ in tasks such as density estimation, time series imputation, and forecasting, showing that it outperforms several strong baselines, including Mambas, ScoreGrad, and Dyffusion.
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。