Research map / Time series / Causal and generative time series
DeepGPD: A Deep Learning Approach for Modeling Geospatio-Temporal Extreme Events
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 |
|---|---|---|---|
| integrate GPD maximum-likelihood loss into the deep model instead of using parameter MSE | DeepGPD · GHCN precipitation test windows | negative log-likelihood: 0.6332± 0.0102 -> 0.5688 ± 0.0036 | Helped |
| use nonlinear spatial modeling and set-valued predictors rather than a linear GPD mapping | DeepGPD · GHCN precipitation test windows | negative log-likelihood: 0.6101 ± 0.0035 -> 0.5688 ± 0.0036 | Helped |
| add GHCN set-valued predictors to NCEP vector-valued predictors (ablation) | DeepGPD · GHCN precipitation test windows | negative log-likelihood: 0.5670 ± 0.0040 -> 0.5688 ± 0.0036 | No clear effect |
| add NCEP vector-valued predictors to GHCN set-valued predictors (ablation) | DeepGPD · GHCN precipitation test windows | negative log-likelihood: 0.5706 ± 0.0035 -> 0.5688 ± 0.0036 | No clear effect |
| use GHCN set-valued predictors only, removing NCEP predictors (ablation) | DeepGPD · GHCN precipitation test windows | negative log-likelihood: 0.5688 ± 0.0036 -> 0.5706 ± 0.0035 | No clear effect |
| use NCEP vector-valued predictors only, removing GHCN predictors (ablation) | DeepGPD · GHCN precipitation test windows | negative log-likelihood: 0.5688 ± 0.0036 -> 0.5670 ± 0.0040 | No clear effect |
Closest papers by meaning
- Quantile-Regression-Ensemble: A Deep Learning Algorithm for Downscaling Extreme Precipitation (AAAI 2024 · similarity 0.83)
- Deep Random Features for Scalable Interpolation of Spatiotemporal Data (ICLR 2025 · similarity 0.83)
- Identifying Spatio-Temporal Drivers of Extreme Events (NeurIPS 2024 · similarity 0.83)
- HR-Extreme: A High-Resolution Dataset for Extreme Weather Forecasting (ICLR 2025 · similarity 0.83)
- GeoPro-Net: Learning Interpretable Spatiotemporal Prediction Models Through Statistically-Guided Geo-Prototyping (AAAI 2025 · similarity 0.82)
- RiverMamba: A State Space Model for Global River Discharge and Flood Forecasting (NeurIPS 2025 · similarity 0.81)
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.