Research map / Time series / Long-term forecasting
Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 global TCN-MF predictions as covariates to the local TCN (ablation) | DeepGLO · electricity, traffic, and wiki rolling prediction, unnormalized | WAPE: 0.092 -> 0.082; 0.169 -> 0.148; 0.212 -> 0.237 | Mixed |
| Combine local TCN with global TCN-MF rather than using global TCN-MF alone (ablation) | DeepGLO · electricity, traffic, and wiki rolling prediction, unnormalized | WAPE: 0.106 -> 0.082; 0.226 -> 0.148; 0.433 -> 0.237 | Helped |
| Add LeveledInit to the local TCN (ablation) | Local TCN · electricity, traffic, and wiki rolling prediction, unnormalized | WAPE: 0.423 -> 0.092; 0.239 -> 0.169; 0.511 -> 0.212 | Helped |
| Use unnormalized rather than normalized training | DeepGLO · electricity, traffic, and PeMSD7(M) rolling prediction | WAPE: 0.133 -> 0.082; 0.166 -> 0.148 | Mixed |
| Replace LeveledInit with a two-network leveling and residual architecture (ablation) | Local TCN · electricity, traffic, and wiki rolling prediction, unnormalized | WAPE: 0.092 -> 0.118; 0.169 -> 0.237; 0.212 -> 0.288 | Mixed |
Closest papers by meaning
- Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts (ICLR 2025 · similarity 0.88)
- Scaling Law for Time Series Forecasting (NeurIPS 2024 · similarity 0.88)
- Unified Training of Universal Time Series Forecasting Transformers (ICML 2024 · similarity 0.86)
- SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series (NeurIPS 2025 · similarity 0.86)
- SIN: Selective and Interpretable Normalization for Long-Term Time Series Forecasting (ICML 2024 · similarity 0.86)
- Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (ICLR 2024 · similarity 0.86)
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.