Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Mitigating Propagation Failures in Physics-informed Neural Networks using Retain-Resample-Release (R3) Sampling
Arka Daw, Jie Bu, Sifan Wang, Paris Perdikaris, Anuj Karpatne
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 R3 sampling instead of uniform dynamic resampling | PINN · Convection β=30, 100k epochs, Nr=1000 | relative L2 error (%): 2.81±1.45 % -> 1.51±0.26 % | Improved |
| add causal loss weighting and causally biased sampling to R3 | R3 · Convection β=30, 300k epochs, Nr=1000 | relative L2 error (%): 0.78±0.18 % -> 0.75±0.12 % | Improved |
| add causal loss weighting and causally biased sampling to R3 | R3 · Convection β=50, 150k epochs, Nr=1000 | relative L2 error (%): 6.03±6.99 % -> 5.99±5.25 % | No clear effect |
| use R3 sampling instead of uniform dynamic resampling | PINN · Convection β=50, 300k epochs, Nr=1000 | relative L2 error (%): 56.9±9.08 % -> 1.98±0.72 % | Improved |
| add causal loss weighting and causally biased sampling to R3 | R3 · Allen Cahn, 200k epochs, Nr=1000 | relative L2 error (%): 0.83±0.15 % -> 0.71±0.007 % | Improved |
Closest papers by meaning
- Efficient Error Certification for Physics-Informed Neural Networks (ICML 2024 · similarity 0.89)
- Sub-Sequential Physics-Informed Learning with State Space Model (ICML 2025 · similarity 0.89)
- Learning from Integral Losses in Physics Informed Neural Networks (ICML 2024 · similarity 0.88)
- PINN Balls: Scaling Second-Order Methods for PINNs with Domain Decomposition and Adaptive Sampling (NeurIPS 2025 · similarity 0.88)
- RoPINN: Region Optimized Physics-Informed Neural Networks (NeurIPS 2024 · similarity 0.87)
- Gradient Alignment in Physics-informed Neural Networks: A Second-Order Optimization Perspective (NeurIPS 2025 · similarity 0.87)
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.
