Research map / Probabilistic and scientific modeling / Differential equations and PDEs
D-CIPHER: Discovery of Closed-form Partial Differential Equations
Krzysztof Kacprzyk, Zhaozhi Qian, Mihaela van der Schaar
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 variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Inhomogeneous heat equation, sigma_R=0.05 | success probability: 0.46 -> 0.64 | Improved |
| replace variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Inhomogeneous heat equation, sigma_R=0.1 | success probability: 0.20 -> 0.42 | Improved |
| replace variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Inhomogeneous heat equation, sigma_R=0.2 | success probability: 0.04 -> 0.12 | Improved |
| replace variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Inhomogeneous heat equation, sigma_R=0.05 | average RMSE: 0.18 -> 0.15 | Improved |
| replace variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Inhomogeneous heat equation, sigma_R=0.2 | average RMSE: 0.27 -> 0.24 | Improved |
| replace variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Forced and damped harmonic oscillator | success probability and RMSE | No clear effect |
| replace variational loss with derivative estimation and MSE loss (ablation) | D-CIPHER · Inhomogeneous wave equation | functional form recovery | Improved |
| compare D-CIPHER with PDE-FIND and WSINDy variants | D-CIPHER · Burgers, homogeneous heat, Kuramoto-Sivashinsky | average RMSE | Improved |
| use CoLLie heuristic instead of CVXOPT convex subproblems | inner optimization · constrained L1 least squares | relative loss error / runtime: 0.005 | Improved |
Closest papers by meaning
- Symmetry-Informed Governing Equation Discovery (NeurIPS 2024 · similarity 0.83)
- Closed-form Solutions: A New Perspective on Solving Differential Equations (ICML 2025 · similarity 0.82)
- Active Symbolic Discovery of Ordinary Differential Equations via Phase Portrait Sketching (AAAI 2025 · similarity 0.82)
- LLM-SR: Scientific Equation Discovery via Programming with Large Language Models (ICLR 2025 · similarity 0.82)
- LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models (ICML 2025 · similarity 0.81)
- ODEFormer: Symbolic Regression of Dynamical Systems with Transformers (ICLR 2024 · similarity 0.80)
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.
