Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Harmonized Cone for Feasible and Non-conflict Directions in Training Physics-Informed Neural Networks
Dohyun Bu, Yujung Byun, Jong-Seok Lee
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 HARMONIC projection to keep update inside H (ablation) | ConFIG · Poisson2d-C | relative L2 error: 0.6856 -> 0.0094 | Improved |
| add HARMONIC projection to keep update inside H (ablation) | ReLoBRaLo · Poisson2d-C | relative L2 error: 0.6602 -> 0.0948 | Improved |
| add HARMONIC projection to keep update inside H (ablation) | CAGrad · Poisson2d-C | relative L2 error: 0.6969 -> 0.0312 | Improved |
| add HARMONIC projection to keep update inside H (ablation) | ConFIG · Wave1d-C | relative L2 error: 0.0668 -> 0.0407 | Improved |
| add HARMONIC projection to keep update inside H (ablation) | ConFIG · Volterra1d | relative L2 error: 0.0018 -> 0.0064 | Hurt |
| add HARMONIC projection to keep update inside H (ablation) | ReLoBRaLo · Volterra1d | relative L2 error: 0.0002 -> 0.0179 | Hurt |
| add HARMONIC projection to keep update inside H (ablation) | ReLoBRaLo · HInv | relative L2 error: 1.5999 -> 0.0467 | Improved |
| add HARMONIC projection to keep update inside H (ablation) | ConFIG · HInv | relative L2 error: 0.0466 -> 0.0558 | Hurt |
| HARMONIC vs best baselines on Poisson2d-C | HARMONIC · Poisson2d-C | relative L2 error: 0.2847 -> 0.0214 | Improved |
| HARMONIC vs ConFIG on Wave1d-C | HARMONIC · Wave1d-C | relative L2 error: 0.0668 -> 0.0655 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ConFIG — ConFIG: Towards Conflict-free Training of Physics Informed Neural Networks (ICLR 2025) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Dual Cone Gradient Descent for Training Physics-Informed Neural Networks (NeurIPS 2024 · similarity 0.91)
- ConFIG: Towards Conflict-free Training of Physics Informed Neural Networks (ICLR 2025 · similarity 0.89)
- Gradient Alignment in Physics-informed Neural Networks: A Second-Order Optimization Perspective (NeurIPS 2025 · similarity 0.89)
- Per-Loss Adapters for Gradient Conflict in Physics-Informed Neural Networks (NeurIPS 2026 · similarity 0.88)
- The Challenges of the Nonlinear Regime for Physics-Informed Neural Networks (NeurIPS 2024 · similarity 0.88)
- Learning a Neural Solver for Parametric PDEs to Enhance Physics-Informed Methods (ICLR 2025 · 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.
