Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Tucker-FNO: Tensor Tucker-Fourier Neural Operator and its Universal Approximation Theory
Guancheng Zhou, Zelin Zeng, Yisi Luo, Qi Xie, Deyu Meng
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 |
|---|---|---|---|
| replace FNO with Tucker-FNO | FNO · Navier-Stokes (T=20) | NMSE: 0.0034 -> 0.0028 | Improved |
| replace D-FNO with Tucker-FNO | D-FNO · Plasticity | NMSE: 0.0097 -> 0.0073 | Improved |
| replace FNO with Tucker-FNO for parameters | FNO · 256x256 PDE data | Params: 593.7 K -> 293.5 K | Improved |
| replace FNO with Tucker-FNO for runtime | FNO · 256x256 PDE data | Time/iter (s): 0.0685 -> 0.0352 | Improved |
| replace OINR-FNO with Tucker-FNO | OINR-FNO · Signal restoration 1024x1024x31 | Params: 1.06 M -> 553 K | Improved |
| replace LRTFR with Tucker-FNO | LRTFR · Video inpainting (SR 0.1) | PSNR (dB): 26.96 -> 28.09 | Improved |
| replace GINO's FNO with Tucker-FNO (12 modes) | GINO · Car-CFD (Fourier mode 12) | Test error (L2): 0.2047 -> 0.2031 | Improved |
| replace GINO's FNO with Tucker-FNO (16 modes) | GINO · Car-CFD (Fourier mode 16) | Test error (L2): 0.1382 -> 0.1297 | Improved |
| increase latent dimension dv from 16 to 64 at k=12 (ablation) | Tucker-FNO · Plasticity | NMSE: 0.0092 -> 0.0064 | Improved |
| increase frequency truncation k from 8 to 16 at dv=32 (ablation) | Tucker-FNO · Plasticity | NMSE: 0.0074 -> 0.0073 | No clear effect |
| increase Tucker rank from 16 to 64 at dv=32 (ablation) | Tucker-FNO · Plasticity | NMSE: 0.0079 -> 0.0071 | Improved |
| increase latent dimension beyond 32 and k beyond 24 (ablation) | Tucker-FNO · Carphone inpainting (SR=0.15) | PSNR | Hurt |
Closest papers by meaning
- Neural Operators with Localized Integral and Differential Kernels (ICML 2024 · similarity 0.89)
- Amortized Fourier Neural Operators (NeurIPS 2024 · similarity 0.89)
- Sensitivity-Constrained Fourier Neural Operators for Forward and Inverse Problems in Parametric Differential Equations (ICLR 2025 · similarity 0.88)
- Guaranteed Approximation Bounds for Mixed-Precision Neural Operators (ICLR 2024 · similarity 0.88)
- Beyond Regular Grids: Fourier-Based Neural Operators on Arbitrary Domains (ICML 2024 · similarity 0.88)
- Component Fourier Neural Operator for Singularly Perturbed Differential Equations (AAAI 2024 · similarity 0.88)
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.
