AI Paper Atlas

Research map / Probabilistic and scientific modeling / Differential equations and PDEs

Deep Equilibrium Based Neural Operators for Steady-State PDEs

Tanya Marwah, Ashwini Pokle, J Zico Kolter, Zachary Chase Lipton, Jianfeng Lu, Andrej Risteski

NeurIPS 2023 · Probabilistic and scientific modeling › Differential equations and PDEs · approach: Neural operators for general geometries · 3 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
weight-tie the FNO block (FNO-WT, 1 block) instead of unique FNO++ layersFNO++ · Darcy Flow cleanrelative L2 error: 0.0064 -> 0.0055Improved
solve for the fixed point with a root solver (FNO-DEQ) instead of unrolling weight-tied layersFNO-WT · Navier-Stokes nu=0.001 cleanrelative L2 error: 0.151 -> 0.128Improved
replace FNO-WT with FNO-DEQFNO-WT · Darcy Flowrelative L2 error: 0.0055 -> 0.0055No clear effect
weight-tied FNO-DEQ (1 block) vs FNO++ with 4 blocks (4x parameters)FNO++ · Navier-Stokes nu=0.01 cleanrelative L2 error: 0.130 -> 0.088Improved
increase FNO depth from 2 to 4 blocksFNO · Darcy Flow cleanrelative L2 error: 0.0105 -> 0.2550Hurt
increase FNO++ depth from 1 to 4 blocksFNO++ · Darcy Flow cleanrelative L2 error: 0.0075 -> 0.0064Improved
increase FNO++ depth from 1 to 4 blocksFNO++ · Navier-Stokes nu=0.001 cleanrelative L2 error: 0.199 -> 0.151Improved
use shallow FNO-DEQ (single FNO layer per block) instead of deep (ablation)FNO-DEQ · Darcy Flow noisy observationsrelative L2 error: 0.0112 -> 0.0136Hurt
use shallow FNO-WT instead of deep (ablation)FNO-WT · Darcy Flow cleanrelative L2 error: 0.0055 -> 0.0057No clear effect
use shallow FNO++ instead of deep FNO++ 1 block (ablation)FNO++ · Darcy Flow noisy observationsrelative L2 error: 0.0145 -> 0.0402Hurt
add noise to inputs (var 0.001) in trainingFNO-DEQ · Navier-Stokes nu=0.001relative L2 error: 0.128 -> 0.144Hurt
increase fixed-point solver steps from 2 to 4 to 8 at Darcy (ablation)FNO-DEQ · Darcy Flow cleantest error: 0.0777 -> 0.00567Improved

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