AI Paper Atlas

Research map / Learning theory and algorithms / Optimal transport

Training neural operators to preserve invariant measures of chaotic attractors

Ruoxi Jiang, Peter Y. Lu, Elena Orlova, Rebecca Willett

NeurIPS 2023 · Learning theory and algorithms › Optimal transport · approach: Neural optimal transport · 4 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 OT loss to RMSE trainingFNO · Lorenz-96 r=0.3L1 histogram error: 0.215 -> 0.057Improved
Add contrastive feature loss to RMSE trainingFNO · Lorenz-96 r=0.3L1 histogram error: 0.215 -> 0.132Improved
Add contrastive feature loss to RMSE trainingFNO · Kuramoto-Sivashinsky r=0.3energy spectrum error: 0.290 -> 0.176Improved
Add OT loss to RMSE trainingFNO · Kuramoto-Sivashinsky r=0.3L1 histogram error: 0.390 -> 0.172Improved
Add contrastive loss at high noiseFNO · Lorenz-96 r=0.3leading Lyapunov exponent error: 0.440 -> 0.064Improved
Add contrastive loss at high noiseFNO · Lorenz-96 r=0.3fractal dimension error: 3.580 -> 1.894Improved
Add OT loss at low noiseFNO · Lorenz-96 r=0.1leading Lyapunov exponent error: 0.013 -> 0.050Hurt
Add OT loss on KS dataFNO · Kuramoto-Sivashinsky r=0.3leading Lyapunov exponent error: 0.101 -> 0.094No clear effect
Reduce OT summary statistics from full to partial (ablation)OT + RMSE · Lorenz-96 r=0.3L1 histogram error: 0.057 -> 0.090Hurt
Reduce OT summary statistics to a non-informative minimum (ablation)OT + RMSE · Lorenz-96 r=0.3L1 histogram error: 0.057 -> 0.221Hurt
Replace Sinkhorn OT with MMD (ablation)OT + RMSE · Kuramoto-Sivashinsky r=0.3L1 histogram error: 0.172 -> 0.245Hurt
Replace RMSE with Sobolev norm and dissipative lossFNO · Lorenz-96 r=0.3L1 histogram error: 0.215 -> 0.246Hurt

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