Stochastic approximation for optimization in shape spaces

In this work, we present a novel approach for solving stochastic shape optimization problems. Our method is the extension of the classical stochastic gradient method to infinite-dimensional shape manifolds. We prove convergence of the method on Riemannian manifolds and then make the connection to shape spaces. The method is demonstrated on a model shape optimization problem from interface identification. Uncertainty arises in the form of a random partial differential equation, where underlying probability distributions of the random coefficients and inputs are assumed to be known. We verify some conditions for convergence for the model problem and demonstrate the method numerically.

[1]  O. Pironneau Optimal Shape Design for Elliptic Systems , 1983 .

[2]  F. Nobile,et al.  Complexity Analysis of stochastic gradient methods for PDE-constrained optimal Control Problems with uncertain parameters , 2018, ESAIM: Mathematical Modelling and Numerical Analysis.

[3]  Daniel Luft,et al.  Computational Investigations of an Obstacle-Type Shape Optimization Problem in the Space of Smooth Shapes , 2019, GSI.

[4]  P. Michor,et al.  The Convenient Setting of Global Analysis , 1997 .

[5]  Orizon Pereira Ferreira,et al.  Kantorovich's Theorem on Newton's Method in Riemannian Manifolds , 2002, J. Complex..

[6]  Martin Bauer,et al.  SOBOLEV METRICS ON SHAPE SPACE, II: WEIGHTED SOBOLEV METRICS AND ALMOST LOCAL METRICS , 2013 .

[7]  V. Schulz,et al.  On Optimization Transfer Operators in Shape Spaces , 2018 .

[8]  Caroline Geiersbach,et al.  Stochastic proximal gradient methods for nonconvex problems in Hilbert spaces , 2021, Comput. Optim. Appl..

[9]  Worst-case shape optimization for the Dirichlet energy , 2016, 1605.05096.

[10]  M. Dambrine,et al.  A first order approach for worst-case shape optimization of the compliance for a mixture in the low contrast regime , 2016 .

[11]  Eldad Haber,et al.  An Effective Method for Parameter Estimation with PDE Constraints with Multiple Right-Hand Sides , 2012, SIAM J. Optim..

[12]  Robert E. Mahony,et al.  Optimization Algorithms on Matrix Manifolds , 2007 .

[13]  L. Eon Bottou Online Learning and Stochastic Approximations , 1998 .

[14]  Dishi Liu,et al.  Quantification of Airfoil Geometry-Induced Aerodynamic Uncertainties - Comparison of Approaches , 2015, SIAM/ASA J. Uncertain. Quantification.

[15]  Martin Bauer,et al.  Sobolev metrics on shape space of surfaces , 2010, 1211.3515.

[16]  Martin Rumpf,et al.  On Shape Optimization with Stochastic Loadings , 2012, Constrained Optimization and Optimal Control for Partial Differential Equations.

[17]  Johannes O. Royset,et al.  Engineering Decisions under Risk Averseness , 2015 .

[18]  Suvrit Sra,et al.  First-order Methods for Geodesically Convex Optimization , 2016, COLT.

[19]  Grégoire Allaire,et al.  A deterministic approximation method in shape optimization under random uncertainties , 2015 .

[20]  Kevin Sturm Minimax Lagrangian Approach to the Differentiability of Nonlinear PDE Constrained Shape Functions Without Saddle Point Assumption , 2015, SIAM J. Control. Optim..

[21]  A Orman,et al.  Optimization of Stochastic Models: The Interface Between Simulation and Optimization , 2012, J. Oper. Res. Soc..

[22]  Martin Bauer Almost local metrics on shape space , 2010 .

[23]  M. Costabel,et al.  Corner Singularities and Analytic Regularity for Linear Elliptic Systems. Part I: Smooth domains. , 2010 .

[24]  Martin Siebenborn,et al.  Computational Comparison of Surface Metrics for PDE Constrained Shape Optimization , 2015, Comput. Methods Appl. Math..

[25]  Michel Métivier,et al.  Semimartingales: A course on stochastic processes , 1986 .

[26]  G. Buttazzo,et al.  Optimal Shapes and Masses, and Optimal Transportation Problems , 2003 .

[27]  Daniel A. Tortorelli,et al.  Topology optimization under uncertainty via non-intrusive polynomial chaos expansion , 2017 .

[28]  Léon Bottou,et al.  On-line learning and stochastic approximations , 1999 .

[29]  D. Mumford,et al.  VANISHING GEODESIC DISTANCE ON SPACES OF SUBMANIFOLDS AND DIFFEOMORPHISMS , 2004, math/0409303.

[30]  Jesús Martínez-Frutos,et al.  Robust shape optimization of continuous structures via the level set method , 2016 .

[31]  Martin Eigel,et al.  An Adaptive Stochastic Galerkin Tensor Train Discretization for Randomly Perturbed Domains , 2020, SIAM/ASA J. Uncertain. Quantification.

[32]  M. C. Delfour,et al.  Shapes and Geometries - Metrics, Analysis, Differential Calculus, and Optimization, Second Edition , 2011, Advances in design and control.

[33]  Helmut Harbrecht,et al.  Shape Optimization for Quadratic Functionals and States with Random Right-Hand Sides , 2015, SIAM J. Control. Optim..

[34]  Martin Siebenborn,et al.  Algorithmic Aspects of Multigrid Methods for Optimization in Shape Spaces , 2016, SIAM J. Sci. Comput..

[35]  Silvere Bonnabel,et al.  Stochastic Gradient Descent on Riemannian Manifolds , 2011, IEEE Transactions on Automatic Control.

[36]  Martin Rumpf,et al.  Shape Optimization Under Uncertainty---A Stochastic Programming Perspective , 2008, SIAM J. Optim..

[37]  Sandra Lowe,et al.  Probability A Graduate Course , 2016 .

[38]  Ohin Kwon,et al.  Magnetic resonance electrical impedance tomography (MREIT): simulation study of J-substitution algorithm , 2002, IEEE Transactions on Biomedical Engineering.

[39]  Thomas Kappeler,et al.  On geodesic exponential maps of the Virasoro group , 2007 .

[40]  Jesús Martínez-Frutos,et al.  Optimal Control of PDEs under Uncertainty: An Introduction with Application to Optimal Shape Design of Structures , 2018 .

[41]  Catherine E. Powell,et al.  An Introduction to Computational Stochastic PDEs , 2014 .

[42]  Dmitriy Drusvyatskiy,et al.  Stochastic Subgradient Method Converges on Tame Functions , 2018, Foundations of Computational Mathematics.

[43]  K. Sturm On shape optimization with non-linear partial differential equations , 2015 .

[44]  Helmut Harbrecht,et al.  The second order perturbation approach for elliptic partial differential equations on random domains , 2018 .

[45]  Martin Siebenborn,et al.  Efficient PDE Constrained Shape Optimization Based on Steklov-Poincaré-Type Metrics , 2015, SIAM J. Optim..

[46]  D. Mumford,et al.  An overview of the Riemannian metrics on spaces of curves using the Hamiltonian approach , 2006, math/0605009.

[47]  John M. Lee Introduction to Riemannian Manifolds , 2019 .

[48]  D. Mumford,et al.  Riemannian Geometries on Spaces of Plane Curves , 2003, math/0312384.

[49]  Martin Bauer,et al.  Overview of the Geometries of Shape Spaces and Diffeomorphism Groups , 2013, Journal of Mathematical Imaging and Vision.

[50]  Volker Schulz,et al.  A Riemannian View on Shape Optimization , 2012, Foundations of Computational Mathematics.

[51]  David Isaacson,et al.  Electrical Impedance Tomography , 1999, SIAM Rev..

[52]  Helmut Harbrecht,et al.  Incorporating knowledge on the measurement noise in electrical impedance tomography , 2019, ESAIM: Control, Optimisation and Calculus of Variations.

[53]  H. Harbrecht,et al.  Solving a Bernoulli type free boundary problem with random diffusion , 2020, ESAIM: Control, Optimisation and Calculus of Variations.

[54]  Kazufumi Ito,et al.  Variational approach to shape derivatives , 2008 .

[55]  J. Zolésio,et al.  Introduction to shape optimization : shape sensitivity analysis , 1992 .

[56]  Caroline Geiersbach,et al.  Projected Stochastic Gradients for Convex Constrained Problems in Hilbert Spaces , 2018, SIAM J. Optim..

[57]  Ulrich Langer,et al.  Shape Optimization of an Electric Motor Subject to Nonlinear Magnetostatics , 2015, SIAM J. Sci. Comput..

[58]  John M. Lee Manifolds and Differential Geometry , 2009 .

[59]  Martin Rumpf,et al.  Stochastic Dominance Constraints in Elastic Shape Optimization , 2016, SIAM J. Control. Optim..

[60]  E. Castorina,et al.  Kantorovich's Theorem on Newton's Method in Riemannian Manifolds , 2002 .

[61]  H. Robbins A Stochastic Approximation Method , 1951 .

[62]  Benedikt Wirth,et al.  Optimization Methods on Riemannian Manifolds and Their Application to Shape Space , 2012, SIAM J. Optim..

[63]  VOLKER H. SCHULZ,et al.  Structured Inverse Modeling in Parabolic Diffusion Problems , 2015, SIAM J. Control. Optim..

[64]  Brigitte Maier,et al.  Fundamentals Of Differential Geometry , 2016 .