Boundary-enhanced time series data imputation with long-term dependency diffusion models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913644421316608 |
|---|---|
| author | Xiao, Chunjing Jiang, Xue Du, Xianghe Yang, Wei Lu, Wei Wang, Xiaomin Chetty, Kevin |
| author_facet | Xiao, Chunjing Jiang, Xue Du, Xianghe Yang, Wei Lu, Wei Wang, Xiaomin Chetty, Kevin |
| contents | Data imputation is crucial for addressing challenges posed by missing values in multivariate time series data across various fields, such as healthcare, traffic, and economics, and has garnered significant attention. Among various methods, diffusion model-based approaches show notable performance improvements. However, existing methods often cause disharmonious boundaries between missing and known regions and overlook long-range dependencies in missing data estimation, leading to suboptimal results. To address these issues, we propose a Diffusion-based time Series Data Imputation (DSDI) framework. We develop a weight-reducing injection strategy that incorporates the predicted values of missing points with reducing weights into the reverse diffusion process to mitigate boundary inconsistencies. Further, we introduce a multi-scale S4-based U-Net, which combines hierarchical information from different levels via multi-resolution integration to capture long-term dependencies. Experimental results demonstrate that our model outperforms existing imputation methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_06585 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Boundary-enhanced time series data imputation with long-term dependency diffusion models Xiao, Chunjing Jiang, Xue Du, Xianghe Yang, Wei Lu, Wei Wang, Xiaomin Chetty, Kevin Machine Learning Social and Information Networks Data imputation is crucial for addressing challenges posed by missing values in multivariate time series data across various fields, such as healthcare, traffic, and economics, and has garnered significant attention. Among various methods, diffusion model-based approaches show notable performance improvements. However, existing methods often cause disharmonious boundaries between missing and known regions and overlook long-range dependencies in missing data estimation, leading to suboptimal results. To address these issues, we propose a Diffusion-based time Series Data Imputation (DSDI) framework. We develop a weight-reducing injection strategy that incorporates the predicted values of missing points with reducing weights into the reverse diffusion process to mitigate boundary inconsistencies. Further, we introduce a multi-scale S4-based U-Net, which combines hierarchical information from different levels via multi-resolution integration to capture long-term dependencies. Experimental results demonstrate that our model outperforms existing imputation methods. |
| title | Boundary-enhanced time series data imputation with long-term dependency diffusion models |
| topic | Machine Learning Social and Information Networks |
| url | https://arxiv.org/abs/2501.06585 |