RDDMPI: Residual Denoising Diffusion Model for Probabilistic Multivariate Time Series Imputation

Abstract

Multivariate time series imputation (MTSI) aims to recover missing values in temporal data composed of multiple interdependent variables. This problem is central to real-world applications such as healthcare monitoring, traffic networks, and energy systems. Recent diffusion-based approaches have shown strong potential for probabilistic imputation by learning to generate missing values through iterative denoising. However, most existing approaches perform diffusion directly in the original data space, requiring the denoising network to simultaneously capture global structure, temporal dynamics, and stochastic variability. This makes the generative task unnecessarily complex, especially when modern deterministic imputers can already provide accurate initial reconstructions. To address this limitation, we propose RDDMPI, a conditional residual diffusion framework that operates directly in residual space. Instead of modeling the full missing signal directly, we reformulate probabilistic imputation as a baseline-residual decomposition, where a pretrained model captures the dominant signal and a diffusion process models the residual uncertainty. To better exploit deterministic guidance, \model{} conditions the reverse denoising process on both the baseline-completed signal and its latent representation, while a reliability-aware conditioning mechanism adaptively controls the influence of baseline information during residual generation. This formulation simplifies the diffusion learning objective, enabling it to focus on structured correction terms rather than reconstructing the full signal. Experiments on multiple benchmark datasets demonstrate that RDDMPI consistently improves both reconstruction accuracy and uncertainty quantification.

Connections between the Föllmer process and the denoising diffusion probabilistic model
Yuta Koike
arXiv2 Sep 2026PreprintarXiv licenceFull textProbability StatisticsarXiv
Spatio-temporal Latent Denoising Diffusion Probabilistic Models for Reduced-order Modeling of Parametrized Dynamical Systems
Michiel Nikken, Nicolò Botteghi, Federico Califano +1
arXiv15 Sep 2026PreprintCC BY 4.0Full textNumerical analysis & scientific computingarXiv
Convergence of Denoising Diffusion Probabilistic Models: A Rényi Divergence Perspective
Wenliang Liu, Santiago Mazuelas, Guanding Yu +3
arXiv18 Oct 2026PreprintCC BY 4.0Full textStatisticsarXiv
Exact Global MCMC with Denoising Diffusion
Mitch Hill
arXiv1 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Denoising as Projection: Constrained Optimization with Gradient-Guided Diffusion
Runyu Zhang, Jiawei Zhang, Gioele Zardini +2
arXiv11 Sep 2026PreprintarXiv licenceFull textOptimisation & controlarXiv

Cite

BibTeX RIS

BibTeX

@misc{jara2026rddmpi,
  title = {{RDDMPI: Residual Denoising Diffusion Model for Probabilistic Multivariate Time Series Imputation}},
  author = {Ramiro Valdes Jara and David Chapman and Adam Meyers},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.11648},
  eprint = {2609.11648},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.11648},
}

RIS

TY  - UNPB
TI  - RDDMPI: Residual Denoising Diffusion Model for Probabilistic Multivariate Time Series Imputation
AU  - Ramiro Valdes Jara
AU  - David Chapman
AU  - Adam Meyers
PY  - 2026
DA  - 2026/09/10
PB  - arXiv
DO  - 10.48550/arxiv.2609.11648
UR  - https://arxiv.org/abs/2609.11648
AB  - Multivariate time series imputation (MTSI) aims to recover missing values in temporal data composed of multiple interdependent variables. This problem is central to real-world applications such as healthcare monitoring, traffic networks, and energy systems. Recent diffusion-based approaches have shown strong potential for probabilistic imputation by learning to generate missing values through iterative denoising. However, most existing approaches perform diffusion directly in the original data space, requiring the denoising network to simultaneously capture global structure, temporal dynamics, and stochastic variability. This makes the generative task unnecessarily complex, especially when modern deterministic imputers can already provide accurate initial reconstructions. To address this limitation, we propose RDDMPI, a conditional residual diffusion framework that operates directly in residual space. Instead of modeling the full missing signal directly, we reformulate probabilistic imputation as a baseline-residual decomposition, where a pretrained model captures the dominant signal and a diffusion process models the residual uncertainty. To better exploit deterministic guidance, \model{} conditions the reverse denoising process on both the baseline-completed signal and its latent representation, while a reliability-aware conditioning mechanism adaptively controls the influence of baseline information during residual generation. This formulation simplifies the diffusion learning objective, enabling it to focus on structured correction terms rather than reconstructing the full signal. Experiments on multiple benchmark datasets demonstrate that RDDMPI consistently improves both reconstruction accuracy and uncertainty quantification.
ER  -

Abstract

Related

Cite