Research map / Time series / Long-term forecasting
Deep Extreme Transformer: Tackling Zero-Inflated Time Series for Precipitation Prediction
Wentao Gao, Xiongren Chen, Xiaojing Du, Wenjun Yu, Andres Mauricio Cifuentes Bernal, Ziqi Xu
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 |
|---|---|---|---|
| use Tweedie loss instead of standard loss (ablation) | DET · South Australia precipitation, 96-step | MSE: 0.650 -> 0.532 | Improved |
| use weighted attention instead of standard attention (ablation) | DET · South Australia precipitation, 96-step | MSE: 0.640 -> 0.532 | Improved |
| use Tweedie mapping instead of standard regression (ablation) | DET · South Australia precipitation, 96-step | MSE: 0.635 -> 0.532 | Improved |
| add weighted attention (ablation) | DET · South Australia precipitation, 96-step | MSE: 0.660 -> 0.532 | Improved |
| add Gaussian perturbation (ablation) | DET · South Australia precipitation, 96-step | MSE: 0.665 -> 0.532 | Improved |
| add Gaussian perturbation (ablation) | DET · South Australia precipitation, 48-step | MSE: 0.635 -> 0.518 | Improved |
| use Tweedie loss instead of standard loss (ablation) | DET · South Australia precipitation, 96-step | MAE: 0.275 -> 0.221 | Improved |
| use DET instead of TimesNet (strongest deep baseline) | TimesNet · South Australia precipitation, 24-step | MSE: 0.652 -> 0.512 | Improved |
| use DET instead of Nonstationary Transformer | Nonstationary Transformer · South Australia precipitation, 24-step | MAE: 0.298 -> 0.212 | Improved |
| use DET instead of zero-inflated statistical models (ZIP, Hurdle) | ZIP · South Australia precipitation, 24-step | MSE: 1.5620 -> 0.5120 | Improved |
| use DET instead of Transformer at long horizon | Transformer · South Australia precipitation, 720-step | MSE: 0.8655 -> 0.5800 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- TimeMixer — TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting (ICLR 2024) · better on 3 of 3 shared results
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 3 of 3 shared results
- TimesNet — TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (ICLR 2023) · better on 3 of 3 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
- Scaling transformer neural networks for skillful and reliable medium-range weather forecasting (NeurIPS 2024 · similarity 0.88)
- OmniCast: A Masked Latent Diffusion Model for Weather Forecasting Across Time Scales (NeurIPS 2025 · similarity 0.86)
- Unified Training of Universal Time Series Forecasting Transformers (ICML 2024 · similarity 0.86)
- Learning from Polar Representation: An Extreme-Adaptive Model for Long-Term Time Series Forecasting (AAAI 2024 · similarity 0.86)
- SAMformer: Unlocking the Potential of Transformers in Time Series Forecasting with Sharpness-Aware Minimization and Channel-Wise Attention (ICML 2024 · similarity 0.86)
- CasCast: Skillful High-resolution Precipitation Nowcasting via Cascaded Modelling (ICML 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.
