PDE Inverse Problems

PDE: Partial Differential Equation

Overview

PDE inverse problems use measured data—often sparse or noisy—to infer unknown coefficients, initial or boundary conditions, sources, or entire physical fields. Classical approaches include regularization, adjoint-based optimization, and Bayesian sampling; newer methods use physics-informed neural networks, neural operators, differentiable solvers, and generative models such as diffusion and flow matching. A central challenge is ill-posedness: a method may reconstruct the observed field well while failing to identify the true parameters or quantify uncertainty reliably. Current work seeks more scalable posterior inference and stronger evaluation of identifiability, robustness to changing observations, and whether predictions support scientific conclusions.

Latest papers

CardsList

114 papers8 papers in the last four weeks, down 56% on the four weeks before. 0.1% of all new papers.

  1. CiGS: Conduction in Gaussian Splats for Interior Temperature Field from Thermal Images

    Nov 15, 2026Liu Zhenyuan, Bharath Seshadri, Bailey Miller +3Inverse ProblemsPDE Inverse Problems

  2. Learning inadequacy of Bayesian neural operators for PDE-constrained inference and design

    Nov 14, 2026Pratyush Kumar Singh, Lianghao Cao, E. Bruce Pitman +1Neural OperatorsPDE Inverse Problems

  3. JAX-WAVE: A Differentiable Physics-Supervised Framework for Inverse Wave Design in Heterogeneous Media

    Nov 10, 2026Honghui Du, Binyao Guo, Yushu PanInverse DesignPDE Inverse Problems

  4. ALETHEIA: A Multi-Frequency Eddy Current Pulsed Thermography Dataset for Neural Operator Learning in Nondestructive Testing

    Nov 8, 2026Changbin Sun, Jie Xiao, Xiaotian Chen +1Neural PDE SolversPDE Operator Learning

  5. Identifying the Shear Piezoelectric Coefficient using Physics-Informed Neural Networks under Exact Boundary and Initial Constraints

    Nov 2, 2026Suhas Suresh BharadwajHard-Constrained PINNsParameter Estimation

  6. Physics-Informed Functional Tucker Method with RKHS Factors for Sparse Spatiotemporal Reconstruction

    Oct 31, 2026Zhengyang Liu, Jianguo Huang, Junxi Hu +7Physics-Informed MLPDE Inverse Problems

  7. PINND: Physics Informed Neural Networks for Solving the Drift-Diffusion Equations

    Oct 23, 2026Luke Ragan, Matteo G. C. Alasio, Enrico Bellotti +1PDE SolvingPDE Inverse Problems

  8. Automatic Differentiation with Constraints in Forward and Inverse Problems

    Oct 20, 2026Nikolay Yavich, Alexander Ryabov, Viacheslav Naumov +4PDE SolvingPDE Inverse Problems

  9. SALTO: Solver-Anchored Lifted Transfer Operators for Nonlinear Event Inference

    Oct 15, 2026Qi Wang, Rylie FosterAdjoint MethodPDE Inverse Problems

  10. Sampling from Neural Operator Priors to Solve Partial Differential Equations

    Oct 12, 2026Andrew Millard, Joshua Murphy, Zheng ZhaoPDE SolvingPartial Differential Equations

  11. An Attention-Guided Graph Neural Network Surrogate for Boundary Integral Computation in Physical Field Reconstruction

    Oct 11, 2026RongZe Gao, YuWei Wu, XinRan Ji +1PDE Surrogate ModelingPDE Inverse Problems

  12. IGA-KAN: Isogeometric Analysis with Physics-Informed Closed-Form Kolmogorov-Arnold Networks for Forward and Inverse PDEs

    Oct 5, 2026Sima Naraghi, Kourosh Parand, Amirhossein Sadr +1Neural PDE SolversKolmogorov-Arnold Networks

  13. OCL-PDE: A Generative Framework for PDE Inverse Problems with Observation-Complementary Latents

    Oct 5, 2026Ding Yang, Chuqi Chen, Chang Ma +1Physics-Informed Generative ModelingPDE Operator Learning

  14. Inferring physical fields in coupled systems with unknown parameters from incomplete observations using physics-constrained attentive neural operators

    Oct 5, 2026Shilun Wei, Xiaoqiang Sun, Wei Li +1Parameter IdentifiabilityPhysics-Informed Neural Operators

  15. Robust Ensemble Guidance for Scientific Inverse Problems

    Oct 4, 2026Zixiang Li, Wei Wang, Yunchao Wei +2Diffusion Model GuidanceDiffusion-Based Inverse Problems

  16. Backward-Consistent Diffusion Sampling for Sparsely Observed PDE Inverse Problems

    Oct 3, 2026Yida Pan, Muhammad H. Ashiq, Chanyong Jung +4PDE Inverse ProblemsDiffusion-Based Inverse Problems

  17. Initial condition recovery in nonlinear damped viscous photoacoustic tomography using a convolutional neural network-guided gradient-free optimization framework

    Oct 1, 2026Madhu Gupta, Anwesa Dey, Prapti Tala +1Medical ImagingGradient-Free Optimization

  18. SCOPE: Observation-Conditioned Full-Target Prediction for Sparse PDE Inference

    Sep 29, 2026Ruichen Xu, Siyao Wang, Fang Wan +8PDE Surrogate ModelingPDE Inverse Problems

  19. PDE-OBS: Controlled Evaluation Across Observation Patterns

    Sep 29, 2026Ruichen Xu, Siyao Wang, Fang Wan +7Benchmark DesignPDE Surrogate Modeling

  20. Beyond Gradient Flow: Identifiability and Recovery from Distribution Snapshots

    Sep 28, 2026Nam D. Nguyen, Valeriya MalyshevaFokker–Planck EquationSystem Identification

  21. PI-NOMT: Physics-Informed Neural Optimal Mass Transport for Brain Fluid Dynamics

    Sep 27, 2026Mehmet Emin Acar, Vahit Bugra Yesilkaynak, Helene Benveniste +1Physics-Informed MLPDE Inverse Problems

  22. NEXT: Physics-Informed Neuro-Spectral Exponential Time Differencing Architectures

    Sep 25, 2026Márcio Marques, Leonardo Mendonça, Leonardo M. Moreira +6Neural PDE SolversPDE Solving

  23. Variational objectives for amortized Bayesian inference in inverse problems: The role of posterior conditioning

    Sep 21, 2026Abhishek Srivastava, Arijit Hazra, Rajesh DubbakuAmortized InferenceBayesian Inverse Problems

  24. PosteriorBench: From Point Estimates to Posterior Matching in Evaluating Generative Inverse Solvers

    Sep 17, 2026Jiachen Yao, Zi-Siang Hsu, Xi Deng +5Uncertainty QuantificationBayesian Inverse Problems

  25. Linearized PINN with pretrained nonlinear layers

    Sep 14, 2026Wenhao Chen, Alexandre M. TartakovskyEfficient Neural Network InferencePDE Solving

  26. Physical-State-Guided Diffusion Sampling for Full-Waveform Inversion

    Sep 11, 2026Chen Min, Haowen Jiang, Zheng Ma +1Physics-Informed Diffusion ModelsSeismic Imaging

  27. Physics-Informed Neural Networks to Infer the Perpendicular Energy Conductivity in the Scrape-Off Layer of Stellarator Devices

    Sep 10, 2026J. Gallego, P. Protopapas, A. Bustos +6Physics-Informed MLPDE Inverse Problems