Research map / Time series / Causal and generative time series
Parameter Decorrelation via Transition-Variance Alignment for Multivariate Time-series Forecasting
Ji-Eun Choi, Jae-Hong Lee, Joon-Hyuk Chang
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 TVA gating of the step size | MLP · Solar | MSE: 0.277 -> 0.234 | Improved |
| add TVA gating of the step size | MLP · Traffic | MSE: 0.508 -> 0.452 | Improved |
| add TVA gating of the step size | MLP · ETTh2 | MSE: 0.374 -> 0.371 | Improved |
| add TVA | MLP · Electricity (T=720) | MSE: 0.274 -> 0.253 | Improved |
| add TVA | iTransformer · Electricity (T=720) | MSE: 0.275 -> 0.244 | Improved |
| add TVA to normalization methods (RevIN, DishTS, SAN) | iTransformer, DLinear, MLP · 8 datasets | MSE: Electricity DishTS iTransformer 0.214 -> 0.247; Solar RevIN iTransformer 0.298 -> 0.276 | Mixed |
| add TVA to Linear | Linear · Solar | MSE: 0.343 -> 0.263 | Improved |
| add TVA to DLinear | DLinear · ETTh1 | MSE: 0.446 -> 0.446 | No clear effect |
| use TVA instead of cosine annealing scheduler (ablation) | MLP · Solar | MSE: 0.275 -> 0.234 | Improved |
| use TVA instead of ReduceLROnPlateau (ablation) | MLP · Traffic | MSE: 0.511 -> 0.452 | Improved |
| use diagonal covariance instead of mean-field scale in TVA (ablation) | MLP+TVA · Solar | MSE: 0.234 -> 0.224 | Improved |
| use diagonal covariance instead of mean-field scale in TVA (ablation) | MLP+TVA · ETTh1 | MSE: 0.438 -> 0.439 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 7 of 9 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.90)
- Unlocking the Power of Patch: Patch-Based MLP for Long-Term Time Series Forecasting (AAAI 2025 · similarity 0.88)
- Multivariate Time Series Forecasting needs Cross Variable Loss (NeurIPS 2026 · similarity 0.88)
- ARM: Refining Multivariate Forecasting with Adaptive Temporal-Contextual Learning (ICLR 2024 · similarity 0.88)
- TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data (AAAI 2025 · similarity 0.87)
- Structured Matrix Basis for Multivariate Time Series Forecasting with Interpretable Dynamics (NeurIPS 2024 · similarity 0.87)
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.
