Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Multiwavelet-based Operator Learning for Differential Equations
Gaurav Gupta, Xiongye Xiao, Paul Bogdan
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 |
|---|---|---|---|
| use careful multiwavelet filter matrices instead of random filters (ablation) | MWT · Darcy Flow s=512 | relative L2 error: 0.2432 -> 0.00654 | Improved |
| use Legendre (uniform measure) basis instead of Chebyshev (ablation) | MWT · KdV s=1024 | relative L2 error: 0.00675 -> 0.00389 | Improved |
| use Legendre basis instead of Chebyshev (ablation) | MWT · Darcy Flow s=512 | relative L2 error: 0.00891 -> 0.00654 | Improved |
| replace FNO with multiwavelet model | operator learner · KdV s=1024 | relative L2 error: 0.0126 -> 0.00389 | Improved |
| replace FNO with multiwavelet model | operator learner · Darcy Flow s=512 | relative L2 error: 0.0106 -> 0.00654 | Improved |
| replace FNO with multiwavelet model at low resolution | operator learner · Darcy Flow s=32 | relative L2 error: 0.0177 -> 0.0152 | Improved |
| train at lower resolution s=256 and test at s=2048 instead of training at s=2048 (ablation) | MWT · Burgers s=2048 | relative L2 error: 0.00189 -> 0.0226 | Hurt |
| use fewer than 3 orthogonal polynomials (k<3) (ablation) | MWT Leg · Euler-Bernoulli beam | relative L2 error across epochs | Hurt |
| increase Fourier modes in FNO vs MWT convolution | FNO · KdV varying fluctuation | relative L2 error | Mixed |
Closest papers by meaning
- Neural Operators with Localized Integral and Differential Kernels (ICML 2024 · similarity 0.87)
- Spectral-Refiner: Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flows (ICLR 2025 · similarity 0.86)
- Toward Efficient Kernel-Based Solvers for Nonlinear PDEs (ICML 2025 · similarity 0.86)
- Physics-Informed Deep Inverse Operator Networks for Solving PDE Inverse Problems (ICLR 2025 · similarity 0.86)
- Generalization Error Bounds for Picard-Type Operator Learning in Nonlinear Parabolic PDEs (NeurIPS 2026 · similarity 0.86)
- Data-Efficient Operator Learning via Unsupervised Pretraining and In-Context Learning (NeurIPS 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.
