Research map / Time series / Long-term forecasting
Goal-Oriented Time-Series Forecasting: Foundation Framework Design
Luca-Andrei Fechete, Mohamed Sana, Fadhel Ayed, Nicola Piovesan, Wenjie Li, Antonio De Domenico, Tareq Si Salem
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 |
|---|---|---|---|
| train with discretized interval policy (DL) instead of baseline policy (ablation) | PatchTST · SynthDS | MAE: 29.1 -> 10.7 | Improved |
| sample all continuous intervals (C-Policy) instead of discretized intervals (ablation) | PatchTST · SynthDS | MAE: 10.7 -> 12.3 | Hurt |
| add averaging patching with finer discretization (D1/2L) over DL (ablation) | PatchTST · SynthDS | MAE: 10.7 -> 11.2 | No clear effect |
| train DL-Policy instead of baseline (ablation) | PatchTST · Traffic | MAE: 1.01 -> 0.62 | Improved |
| train DL-Policy instead of baseline (ablation) | PatchTST · BLW-TrafficDS | MAE: 50.7 -> 22.4 | Improved |
| train DL-Policy instead of baseline (ablation) | PatchTST · Weather | MAE: 0.37 -> 0.20 | Improved |
| train DL-Policy instead of baseline (ablation) | PatchTST · Electricity | MAE: 1.39 -> 1.29 (DL better) but patching 1.75 and 1.81 worse than B | Mixed |
| train patching policies instead of baseline (ablation) | iTransformer · Electricity | MAE: 1.39 -> 4.09 / 4.21 | Hurt |
| use patching D1/2L instead of DL (ablation) | PatchTST · Traffic | MAE: 0.62 -> 0.59 | No clear effect |
| use C-Policy instead of discretized policies (ablation) | DLinear · real-world traces | MAE | Mixed |
| apply goal-oriented forecasting to wireless energy saving | proposed method · wireless network energy-saving | energy-saving error: 0.950 kW -> 337 W | Improved |
Closest papers by meaning
- ProbTS: Benchmarking Point and Distributional Forecasting across Diverse Prediction Horizons (NeurIPS 2024 · similarity 0.88)
- Improving Time Series Forecasting via Instance-aware Post-hoc Revision (NeurIPS 2025 · similarity 0.88)
- TimePerceiver: An Encoder-Decoder Framework for Generalized Time-Series Forecasting (NeurIPS 2025 · similarity 0.87)
- fev-bench: A Realistic Benchmark for Time Series Forecasting (NeurIPS 2026 · similarity 0.87)
- An Analysis of Linear Time Series Forecasting Models (ICML 2024 · similarity 0.87)
- Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (ICLR 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.
