AI Paper Atlas

Research map / Image generation / GANs

Probabilistic Forecasting with Generative Networks via Scoring Rule Minimization

Lorenzo Pacchiardi, Rilwan A. Adewoyin, Peter Dueben, Ritabrata Dutta

JMLR 2024 · Image generation › GANs · approach: Generative adversarial network theory · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
train with Energy-Kernel Score instead of WGAN-GP (3)generative forecaster · Lorenz96calibration error: 0.1648 -> 0.0104Improved
train with Energy Score instead of GAN (3)generative forecaster · Lorenz63NRMSE: 0.0698 -> 0.0105Improved
train with Kernel Score instead of regression lossGRU forecaster · Lorenz96NRMSE: 0.0198 -> 0.0164Improved
train with Kernel Score instead of regression lossGRU forecaster · Lorenz63NRMSE: 0.0079 -> 0.0083No clear effect
use Kernel Score instead of Energy ScoreGRU forecaster · Lorenz96calibration error: 0.0205 -> 0.2196Hurt
use Energy-Kernel Score instead of Energy ScoreGRU forecaster · Lorenz63calibration error: 0.0380 -> 0.1000Hurt
pick GAN config (2) for calibration instead of (1) for deterministic skillGAN · Lorenz63NRMSE: 0.0274 -> 0.2425Hurt
pick GAN config (1) for deterministic skill instead of (2)GAN · Lorenz63calibration error: 0.0860 -> 0.4830Hurt
use Patched Energy Score (patch 8) instead of Energy ScoreU-Net forecaster · WeatherBench Z500calibration error: 0.0863 -> 0.0550Improved
add patching to Energy Score (patch 16 vs unpatched)U-Net forecaster · WeatherBench Z500NRMSE: 0.1208 -> 0.1186Improved
combine Kernel Score with Variogram Score instead of Kernel aloneU-Net forecaster · WeatherBench Z500calibration error: 0.0797 -> 0.1704Hurt
combine Energy with Kernel Score instead of Energy aloneU-Net forecaster · WeatherBench Z500NRMSE: 0.1208 -> 0.1194Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy