DAMR: Dynamic Adjacency Matrix Representation Learning for Multivariate Time Series Imputation
Summary: Dynamic adjacency matrices capture evolving spatial correlations in multivariate time series, decomposing into constant, long-term trends and periodic patterns. DAMR aggregates these dynamic graphs and applies a graph representation learning layer for missing-value imputation, yielding up to 19.4% MAE improvement over SOTA. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Xiaobin Ren
- 2. Kaiqi Zhao
- 3. Patricia Riddle
- 4. Katerina Taškova
- 5. Lianyan Li
- 6. Qingyi Pan
Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 6,900 | Kamel: A Scalable BERT-based System for Trajectory Imputation | 2024 | VLDB | 4.8925595e-05 |
| 9,242 | ImputeVIS: An Interactive Evaluator to Benchmark Imputation Techniques for Time Series Data | 2024 | VLDB | 4.3690661e-05 |
| 10,744 | DIM-SUM: Dynamic IMputation for Smart Utility Management | 2025 | VLDB | 4.1945683e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 0 of 0 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next