Research map / Time series / Long-term forecasting
HN-MVTS: HyperNetwork-based Multivariate Time Series Forecasting
Andrey Savchenko, Oleg Kachan
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 HN-MVTS hypernetwork | DLinear · Weather | MSE: 0.1733 -> 0.1425 (H=96) | Improved |
| add HN-MVTS hypernetwork | TSMixer · ECL | MSE: 0.1972 -> 0.1666 (H=192) | Improved |
| add HN-MVTS hypernetwork | PatchTST · PEMS07 | MSE: 0.1619 -> 0.1415 (H=336) | Improved |
| add HN-MVTS hypernetwork | iTransformer · PEMS08 (H=336) | MSE: 0.1770 -> 0.1769 (H=336) | No clear effect |
| add HN-MVTS hypernetwork | iTransformer · ETTm1 | MSE: 0.3034 -> 0.3171 (H=96) | Hurt |
| add HN-MVTS hypernetwork | PatchTST · ETTm1/ETTm2 (short horizons) | MSE: 0.2528 -> 0.2603 (ETTm1 H=48) | No clear effect |
| add HN-MVTS hypernetwork | ModernTCN · ETTm2 / PEMS03 / PEMS07 | MSE: 0.2857 -> 0.3075 (ETTm2 H=336); 0.1313 -> 0.1178 (PEMS03 H=48) | Mixed |
| add HN-MVTS hypernetwork | DLinear · ECL | training time per epoch (s): 15.31 -> 17.24 (H=48) | Hurt |
| add HN-MVTS hypernetwork | PatchTST · ECL | training time per epoch (s): 52.38 -> 66.26 (H=336) | Hurt |
| initialize embeddings from correlations instead of random (ablation) | HN-MVTS · benchmarks | MSE | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 1 of 2 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
- HyperIMTS: Hypergraph Neural Network for Irregular Multivariate Time Series Forecasting (ICML 2025 · similarity 0.91)
- HDT: Hierarchical Discrete Transformer for Multivariate Time Series Forecasting (AAAI 2025 · similarity 0.89)
- TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data (AAAI 2025 · similarity 0.89)
- HyperMixer: Specializable Hypergraph Channel Mixing for Long-term Multivariate Time Series Forecasting (AAAI 2025 · similarity 0.88)
- Hi-Patch: Hierarchical Patch GNN for Irregular Multivariate Time Series (ICML 2025 · similarity 0.88)
- Hierarchical Classification Auxiliary Network for Time Series Forecasting (AAAI 2025 · similarity 0.88)
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.
