Research map / Time series / Long-term forecasting
Beyond Missing Data Imputation: Information-Theoretic Coupling of Missingness and Class Imbalance for Optimal Irregular Time Series Classification
Xin Qin, Mengna Liu, Wenjie Wang, Shuxin Li, Tianjiao Li, Xiufeng Liu, Xu Cheng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add MPE (missingness pattern encoder) (ablation) | SPECTRA · P12 | AUPR: 50.7 -> 54.4 | Improved |
| add MPE (missingness pattern encoder) (ablation) | SPECTRA · P19 | AUPR: 51.0 -> 59.4 | Improved |
| add CGFR (prototype classification) (ablation) | SPECTRA · P12 | AUPR: 44.6 -> 54.4 | Improved |
| add CGFR (prototype classification) (ablation) | SPECTRA · P19 | AUPR: 49.6 -> 59.4 | Improved |
| add MAFF (missingness-aware frequency filtering) (ablation) | SPECTRA · PAM | F1: 86.9 -> 98.0 | Improved |
| add MAFF (missingness-aware frequency filtering) (ablation) | SPECTRA · P19 | AUPR: 54.7 -> 59.4 | Improved |
| add SCSE (self-calibrating spectral enhancement) (ablation) | SPECTRA · P19 | AUPR: 58.5 -> 59.4 | No clear effect |
| add AG-DRR (dynamic receptive field) (ablation) | SPECTRA · P19 | AUPR: 58.1 -> 59.4 | No clear effect |
| add MPE (missingness pattern encoder) (ablation) | SPECTRA · PAM | F1: 97.9 -> 98.0 | No clear effect |
| replace Focal/LDAM/CB imbalance losses with CGFR prototype objective (ablation) | SPECTRA · P19 | AUPR: 58.3 -> 59.4 | Improved |
| replace Focal/LDAM/CB imbalance losses with CGFR prototype objective (ablation) | SPECTRA · P12 | AUPR: 52.2 -> 54.4 | Improved |
| use SPECTRA instead of MTSFormer | SPECTRA · P19 | AUPR: 57.7 -> 59.4 | Improved |
Closest papers by meaning
- Frequency-aware Generative Models for Multivariate Time Series Imputation (NeurIPS 2024 · similarity 0.86)
- Discretizing Continuous Time Series for Imputation with Masked Diffusion Training (NeurIPS 2026 · similarity 0.85)
- LSCD: Lomb--Scargle Conditioned Diffusion for Time series Imputation (ICML 2025 · similarity 0.85)
- HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation (NeurIPS 2026 · similarity 0.85)
- Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning (AAAI 2025 · similarity 0.85)
- Content-aware Balanced Spectrum Encoding in Masked Modeling for Time Series Classification (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
