<ul><li>Physics-informed neural networks (PINNs) are powerful for solving problems involving partial differential equations (PDEs) by incorporating physical laws.</li><li>A new iterative multi-objective PINN ensemble Kalman filter (MoPINNEnKF) framework is proposed to improve the robustness and accuracy of PINNs in forward and inverse problems.</li><li>The framework uses the ensemble Kalman filter and the non-dominated sorting genetic algorithm III (NSGA-III) to refine data loss components and update PINNs' parameters iteratively.</li><li>Numerical results on benchmark problems show that MoPINNEnKF outperforms standard PINNs in handling noisy data and missing physics.</li></ul>

MoPINNEnKF: Iterative Model Inference using generic-PINN-based ensemble Kalman filter

Discover more