Research map / Time series / Causal and generative time series
Not All Frequencies Are Equal: Energy-Adaptive Diffusion for Time Series Forecasting
Zining Qin, Huiling Qin, Chenhao Wang, Jianxiong Guo, Tian Wang, Weijia Jia
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 energy-adaptive noise scheduler (ablation) | EADiff · Beijing Air (H=96) | RMSE: 1.028 -> 0.681 | Improved |
| add energy-adaptive noise scheduler (ablation) | EADiff · Exchange (H=96) | RMSE: 0.840 -> 0.500 | Improved |
| add energy-adaptive noise scheduler (ablation) | EADiff · ETTh1 (H=96) | RMSE: 1.137 -> 0.837 | Improved |
| add energy-adaptive noise scheduler (ablation) | EADiff · ETTh2 (H=96) | RMSE: 0.558 -> 0.538 | Improved |
| add wavelet decomposition (ablation) | EADiff · Exchange (H=96) | RMSE: 1.301 -> 0.500 | Improved |
| add wavelet decomposition (ablation) | EADiff · ETTh1 (H=96) | RMSE: 2.742 -> 0.837 | Improved |
| add wavelet decomposition (ablation) | EADiff · Beijing Air (H=96) | RMSE: 0.682 -> 0.681 | No clear effect |
| use decomposition level 1 instead of level 5 (ablation) | EADiff · Beijing Air (H=96) | RMSE: 1.106 -> 0.681 | Improved |
| use decomposition level 5 instead of level 1 (ablation) | EADiff · Exchange (H=96) | RMSE: 1.397 -> 0.500 | Improved |
| use sym2 instead of bior3.1 wavelet (ablation) | EADiff · Beijing Air (H=96) | RMSE: 6.866 -> 1.106 | Improved |
| use db1 (L6) instead of sym2 (L5) wavelet (ablation) | EADiff · ETTh1 (H=96) | RMSE: 1.031 -> 0.837 | Mixed |
Closest papers by meaning
- ANT: Adaptive Noise Schedule for Time Series Diffusion Models (NeurIPS 2024 · similarity 0.90)
- Non-stationary Diffusion For Probabilistic Time Series Forecasting (ICML 2025 · similarity 0.89)
- DiffPTS: Rethinking Diffusion ELBO for Probabilistic Time Series Forecasting (NeurIPS 2026 · similarity 0.89)
- Conditional Diffusion Model with Nonlinear Data Transformation for Time Series Forecasting (ICML 2025 · similarity 0.89)
- Auto-Regressive Moving Diffusion Models for Time Series Forecasting (AAAI 2025 · similarity 0.89)
- A Non-isotropic Time Series Diffusion Model with Moving Average Transitions (ICML 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.
