Truncated Nonsmooth Newton Multigrid Methods for Simplex-Constrained Minimization Problems
نویسندگان
چکیده
We present a multigrid method for the minimization of strongly convex functionals defined on a finite product of simplices. Such problems result, for example, from the discretization of multi-component phase-field problems. Our algorithm is globally convergent, requires no regularization parameters, and achieves multigrid convergence rates. We present numerical results for the vector-valued Allen–Cahn equation and observe that the convergence rate is independent from the temperature parameter and the number of components. AMS classification: 65K15, 90C25, 49M20
منابع مشابه
A nonsmooth Newton multigrid method for a hybrid, shallow model of marine ice sheets
The time evolution of ice sheets and ice shelves is modelled by combining a shallow lubrication approximation for shear deformation with the shallow shelf approximation for basal sliding, along with the mass conservation principle. At each time step two p-Laplace problems and one transport problem are solved. Both p-Laplace problems are formulated as minimisation problems. They are approximated...
متن کاملGlobally Convergent Multigrid Methods for Porous Medium Type Problems
We consider the fast solution of large, piecewise smooth minimization problems as typically arising from the nite element discretization of porous media ow. For lack of smoothness, usual Newton multigrid methods cannot be applied. We propose a new approach based on a combination of convex minization with constrained Newton linearization. No regularization is involved. We show global convergence...
متن کاملSpeeding-Up Convergence via Sequential Subspace Optimization: Current State and Future Directions
This is an overview paper written in style of research proposal. In recent years we introduced a general framework for large-scale unconstrained optimization – Sequential Subspace Optimization (SESOP) and demonstrated its usefulness for sparsity-based signal/image denoising, deconvolution, compressive sensing, computed tomography, diffraction imaging, support vector machines. We explored its co...
متن کاملOn constrained Newton linearization and multigrid for variational inequalities
Abstra t. We onsider the fast solution of a lass of large, pie ewise smooth minimization problems. For la k of smoothness, usual Newton multigrid methods annot be applied. We propose a new approa h based on a ombination of onvex minization with onstrained Newton linearization. No regularization is involved. We show global onvergen e of the resulting monotone multigrid methods and give polylogar...
متن کاملAn efficient one-layer recurrent neural network for solving a class of nonsmooth optimization problems
Constrained optimization problems have a wide range of applications in science, economics, and engineering. In this paper, a neural network model is proposed to solve a class of nonsmooth constrained optimization problems with a nonsmooth convex objective function subject to nonlinear inequality and affine equality constraints. It is a one-layer non-penalty recurrent neural network based on the...
متن کامل