Diffusion-based Time Series Imputation and Forecasting with Structured State Space Models

Abstract

The imputation of missing values represents a significant obstacle for many real-world data analysis pipelines. Here, we focus on time series data and put forward SSSD, an imputation model that relies on two emerging technologies, (conditional) diffusion models as state-of-the-art generative models and structured state space models as internal model architecture, which are particularly suited to capture long-term dependencies in time series data. We demonstrate that SSSD matches or even exceeds state-of-the-art probabilistic imputation and forecasting performance on a broad range of data sets and different missingness scenarios, including the challenging blackout-missing scenarios, where prior approaches failed to provide meaningful results.

V3: Updated results for the solar dataset. 36 pages, 13 figures. Version published by Transactions on Machine Learning Research in 2022 (TMLR ISSN 2835-8856) https://openreview.net/forum?id=hHiIbk7ApW. Source code under https://github.com/AI4HealthUOL/SSSD

Cited by in corpus (6)

• Generative AI for Synthetic Data Across Multiple Medical Modalities: A Systematic Review of Recent Developments and Challenges
• SpectralDiff: A Generative Framework for Hyperspectral Image Classification with Diffusion Models
• Boundary-enhanced time series data imputation with long-term dependency diffusion models
• 4D Facial Expression Diffusion Model
• CoSTI: Consistency Models for (a faster) Spatio-Temporal Imputation
• Diffusion models as probabilistic neural operators for recovering unobserved states of dynamical systems