Get Started
Home
Topics
Search
Library
Research questionHow can physics-informed neural networks handle stiff, multiscale ion-electronic transport while supporting inverse estimation?Ion-electronic drift-diffusion in oxide heterostructures combines sharply different spatial scales with severe numerical stiffness, making standard physics-informed neural networks difficult to train reliably. A useful surrogate must also preserve the governing electrostatics well enough for inverse estimation.
AI
Machine Learning
Neural and Evolutionary Computing
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.Physics-Informed Neural Network Surrogate for Oxygen Vacancy Dynamics in epitaxial $\mathrm{SrTiO_3}$ on Si memristors via Dynamic Spectral OptimizationThe evidence concerns oxygen-vacancy and ion-electronic transport in a Pt/SrTiO₃/Si memristive heterostructure with a 20 nm SrTiO₃ film and 380 μm Si substrate. The reported surrogate reproduces conductive-AFM current-voltage hysteresis with R² > 0.96, maintains continuous-space Poisson consistency, and supports differentiable inverse parameter estimation; the input does not establish broader device or material generalization.research paper · Sep 2, 2026
Related questions
How can chemical transport models learn unknown constitutive laws and initial conditions from experimental data?How can surrogate models accelerate TCAD exploration of large 3D FinFET designs without losing physics across geometries?How can neural operators learn nonequilibrium PDE dynamics while preserving energy conservation and entropy production?How can compact device surrogates stay physically consistent and accurate across unseen process or geometry splits with scarce data?