Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Guiding continuous operator learning through Physics-based boundary constraints
Nadim Saad, Gaurav Gupta, Shima Alizadeh, Danielle C. Maddix
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add BOON Dirichlet/Neumann/periodic kernel correction to FNO | FNO · Burgers' Dirichlet single-step (nu=0.02) | relative L2 error: 0.0028 -> 0.000084 | Improved |
| add BOON kernel correction to FNO | FNO · Stokes' Dirichlet multi-step, nu in {0.1..0.001} | relative L2 error: e.g. 0.0099 -> 0.0089 (nu=0.1); 0.0273 -> 0.0183 (nu=0.001) | Improved |
| add BOON kernel correction to FNO | FNO · Heat equation Neumann multi-step, N in {50,100,250,500} | relative L2 error: N=50: 0.00915 -> 0.00675; N=100: 0.00845 -> 0.00501; N=250: 0.00925 -> 0.00893; N=500: 0.01239 -> 0.01187 | Mixed |
| add BOON kernel correction to FNO | FNO · Wave equation Neumann multi-step, N in {25,50,100} | relative L2 error: N=25: 0.00140 -> 0.00097; N=50: 0.00093 -> 0.000893; N=100: 0.00119 -> 0.00096 | Mixed |
| add periodic kernel correction to FNO | FNO · Burgers' periodic multi-step, N in {32,64,128} | relative L2 error: N=32: 0.00133 -> 0.00097; N=64: 0.00104 -> 0.00093; N=128: 0.00201 -> 0.00101 | Mixed |
| add BOON kernel correction to FNO | FNO · Navier-Stokes lid cavity multi-step, Re in {10,100,1000} | relative L2 error: Re=10: 0.00249 -> 0.00213; Re=100: 0.00186 -> 0.00143; Re=1000: 0.00215 -> 0.00176 | Improved |
Closest papers by meaning
- BENO: Boundary-embedded Neural Operators for Elliptic PDEs (ICLR 2024 · similarity 0.88)
- Physics-Informed Deep Inverse Operator Networks for Solving PDE Inverse Problems (ICLR 2025 · similarity 0.87)
- RIGNO: A Graph-based Framework For Robust And Accurate Operator Learning For PDEs On Arbitrary Domains (NeurIPS 2025 · similarity 0.86)
- Solving Differential Equations with Constrained Learning (ICLR 2025 · similarity 0.86)
- Using Uncertainty Quantification to Characterize and Improve Out-of-Domain Learning for PDEs (ICML 2024 · similarity 0.86)
- Reference Neural Operators: Learning the Smooth Dependence of Solutions of PDEs on Geometric Deformations (ICML 2024 · similarity 0.86)
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.
