CARE-SAV: A Conditioning-Aware Random-Feature Framework for Energy-Stable Simulation of Gradient Flows

DOI10.48550/arXiv.2608.25300MaRDI QIDQ7284979arXivFDO

Authors Bingcheng Hu, Zhaoxiang Li

Publication date 26 August 2026

Published in arXiv

Tldr We develop a Conditioning-Aware Representation Enhancement with Scalar Auxiliary Variable (CARE-SAV) framework to construct a compact spatial approximation space from flexible candidate features and evolve the gradient-flow dynamics directly within this space, preserving the discrete energy-dissipation law while providing flexibility in discretization.

Full work available at URL https://doi.org/10.48550/arXiv.2608.25300

Mathematics Subject Classification ID

Numerical radial basis function approximation (65D12) Numerical computation of matrix norms, conditioning, scaling (65F35) Stability and convergence of numerical methods for initial value and initial-boundary value problems involving PDEs (65M12) Finite element, Rayleigh-Ritz and Galerkin methods for initial value and initial-boundary value problems involving PDEs (65M60)

arXiv classification

math.NA

Cites work

• A Highly Efficient and Accurate New Scalar Auxiliary Variable Approach for Gradient Flows
• A new variable shape parameter strategy for RBF approximation using neural networks
• A phase-field model and its numerical approximation for two-phase incompressible flows with different densities and viscosities
• A variant of scalar auxiliary variable approaches for gradient flows
• An energy-stable and convergent finite-difference scheme for the phase field crystal equation
• Bridging traditional and machine learning-based algorithms for solving PDEs: the random feature method
• Characterizing the stabilization size for semi-implicit Fourier-spectral method to phase field equations
• Convergence analysis of a second order convex splitting scheme for the modified phase field crystal equation
• Convergence and error analysis for the scalar auxiliary variable (SAV) schemes to gradient flows
• Decoupled energy stable schemes for phase-field models of two-phase complex fluids
• Decoupled, linear, and energy stable finite element method for the Cahn-Hilliard-Navier-Stokes-Darcy phase field model
• DGM: a deep learning algorithm for solving partial differential equations
• Dissipative numerical schemes on Riemannian manifolds with applications to gradient flows
• Efficient energy stable numerical schemes for a phase field moving contact line model
• Efficient second order unconditionally stable schemes for a phase field moving contact line model using an invariant energy quadratization approach
• Element‐free Galerkin methods
• Energy Dissipation Preserving Feature-based DNN Galerkin Methods for Gradient Flows
• Error analysis of stabilized semi-implicit method of Allen-Cahn equation
• Extreme learning machine collocation for the numerical solution of elliptic PDEs with sharp gradients
• Fourth-Order Time-Stepping for Stiff PDEs
• Linearly first- and second-order, unconditionally energy stable schemes for the phase field crystal model
• Local Extreme Learning Machines and Domain Decomposition for Solving Linear and Nonlinear Partial Differential Equations
• Local Randomized Neural Networks with Discontinuous Galerkin Methods for Partial Differential Equations
• Multiple scalar auxiliary variable (MSAV) approach and its application to the phase-field vesicle membrane model
• Numerical analysis of second order, fully discrete energy stable schemes for phase field models of two-phase incompressible flows
• Numerical approximations for a phase field dendritic crystal growth model based on the invariant energy quadratization approach
• Numerical approximations for the molecular beam epitaxial growth model based on the invariant energy quadratization method
• On energy dissipation theory and numerical stability for time-fractional phase-field equations
• Optimal error estimates for the scalar auxiliary variable finite-element schemes for gradient flows
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Provably unconditionally stable, second-order time-accurate, mixed variational methods for phase-field models
• Randomized neural network with Petrov-Galerkin methods for solving linear and nonlinear partial differential equations
• Second Order Fully Discrete Energy Stable Methods on Staggered Grids for Hydrodynamic Phase Field Models of Binary Viscous Fluids
• Solving partial differential equation based on extreme learning machine
• Stability and error estimates of the SAV Fourier-spectral method for the phase field crystal equation
• The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems
• The random feature method for solving interface problems
• The Random Feature Method for Time-Dependent Problems
• The scalar auxiliary variable (SAV) approach for gradient flows

This page was built for publication: CARE-SAV: A Conditioning-Aware Random-Feature Framework for Energy-Stable Simulation of Gradient Flows

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7284979)