AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Embed and Emulate: Learning to estimate parameters of dynamical systems with uncertainty quantification

Ruoxi Jiang, Rebecca Willett

NeurIPS 2022 · Probabilistic and scientific modeling › Dynamical systems · approach: State-space inference for stochastic systems · 1 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
add Embed & Emulate learned embedding and emulator to EnKI instead of Runge-Kutta solver with moment vectorEnKI · Lorenz-96 n=1000MAPE (%) for F: 15.48 -> 6.59Improved
replace Runge-Kutta solver by learned emulator in EnKIEnKI · Lorenz-96 n=1000total time (min): 8,000 -> 95.0Improved
train Embed & Emulate with contrastive objective instead of supervised regression (n=1000)regression head · Lorenz-96 n=1000MAPE (%) for c: 11.46 -> 9.34Improved
use Embed & Emulate instead of supervised regression with fewer training samples (n=500)regression head · Lorenz-96 n=500MAPE (%) for F: 11.97 -> 11.19Improved
increase training size from 500 to 1000 (ablation)Embed & Emulate regression head · Lorenz-96MAPE (%) for F: 11.19 -> 6.30Improved
use EnKI rather than regression head for uncertainty-aware estimates (ablation)Embed & Emulate · Lorenz-96 n=1000MAPE (%) for F: 6.30 -> 6.59No clear effect
use EnKI posterior rather than regression head point estimates (ablation)Embed & Emulate · Lorenz-96 n=1000CRPS for F: 0.412 -> 0.360Improved
use Embed & Emulate + EnKI instead of NPE-CEnKI · Lorenz-96 n=1000CRPS for F: 0.593 -> 0.360Improved
use Embed & Emulate + EnKI instead of SNL+EnKI · Lorenz-96 n=1000CRPS for F: 1.304 -> 0.360Improved
remove the regression head loss (ablation)Embed & Emulate · Lorenz-96 n=4000MAPE (%) for F: 8.39 -> 16.75Hurt
add empirical Bayes prior from regression head (ablation)Embed & Emulate · Lorenz-96 n=4000MAPE (%) for F: 8.39 -> 3.39Improved
use learned Embed & Emulate under observation noise r=0.1, 0.2EnKI with Embed & Emulate · Lorenz-96 noisy observationsposterior consistencyImproved

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