New publication for MIPORE!

Milad, Giovanni, Monica, and Alberto have published a new paper in PNAS Nexus. The study tackles the challenge of quantifying parametric uncertainty in partial differential equations by introducing a physics‑informed neural network that acts as a differentiable solver, learning the full solution family of steady‑state Darcy flow in a single training run. The method delivers accurate flow solutions and enables efficient uncertainty quantification, offering a versatile framework for physics‑constrained, data‑driven modeling of heterogeneous systems.

Kudos to the team!