Newton Hard-Thresholding Pursuit for Sparse Linear Complementarity Problem via A New Merit Function

Solutions to the linear complementarity problem (LCP) are naturally sparse in many applications such as bimatrix games and portfolio section problems. Despite that it gives rise to the hardness, sparsity makes optimization faster and enables relatively large scale computation. Motivated by this, we take the sparse LCP into consideration, investigating the existence and boundedness of its solution set as well as introducing a new merit function, which allows us to convert the problem into a sparsity constrained optimization. The function turns out to be continuously differentiable and twice continuously differentiable for some chosen parameters. Interestingly, it is also convex if the involved matrix is positive semidefinite. We then explore the relationship between the solution set to the sparse LCP and stationary points of the sparsity constrained optimization. Finally, Newton hard thresholding pursuit is adopted to solve the sparsity constrained model. Numerical experiments demonstrate that the problem can be efficiently solved through the new merit function.

[1]  Philip Wolfe,et al.  An algorithm for quadratic programming , 1956 .

[2]  Shenglong Zhou,et al.  Extragradient thresholding methods for sparse solutions of co-coercive NCPs , 2015 .

[3]  Richard W. Cottle,et al.  Linear Complementarity Problem. , 1992 .

[4]  Naihua Xiu,et al.  Global and Quadratic Convergence of Newton Hard-Thresholding Pursuit , 2019, J. Mach. Learn. Res..

[5]  M. Fukushima,et al.  On stationary points of the implicit Lagrangian for nonlinear complementarity problems , 1995 .

[6]  Jein-Shan Chen,et al.  A family of NCP functions and a descent method for the nonlinear complementarity problem , 2008, Comput. Optim. Appl..

[7]  Jein-Shan Chen,et al.  On the generalized Fischer-Burmeister merit function for the second-order cone complementarity problem , 2013, Math. Comput..

[8]  Francisco Facchinei,et al.  A nonsmooth inexact Newton method for the solution of large-scale nonlinear complementarity problems , 1997, Math. Program..

[9]  Jiang Xie,et al.  Randomized Portfolio Selection , with Constraints , 2007 .

[10]  Xiao-Hong Liu,et al.  Coerciveness of some merit functions over symmetric cones , 2009 .

[11]  F. Clarke Generalized gradients and applications , 1975 .

[12]  A. Fischer An NCP–Function and its Use for the Solution of Complementarity Problems , 1995 .

[13]  Olvi L. Mangasarian,et al.  Nonlinear complementarity as unconstrained and constrained minimization , 1993, Math. Program..

[14]  M. Fiedler,et al.  On matrices with non-positive off-diagonal elements and positive principal minors , 1962 .

[15]  Xiaojun Chen,et al.  A penalized Fischer-Burmeister NCP-function , 2000, Math. Program..

[16]  Naihua Xiu,et al.  Optimality conditions for sparse nonlinear programming , 2017 .

[17]  Chao Zhang,et al.  Minimal Zero Norm Solutions of Linear Complementarity Problems , 2014, J. Optim. Theory Appl..

[18]  Michael C. Ferris,et al.  Complementarity: Applications, Algorithms and Extensions , 2010 .

[19]  R. Stephenson A and V , 1962, The British journal of ophthalmology.

[20]  J. Hiriart-Urruty,et al.  Generalized Hessian matrix and second-order optimality conditions for problems withC1,1 data , 1984 .

[21]  Masao Fukushima,et al.  A globally convergent Newton method for solving strongly monotone variational inequalities , 1993, Math. Program..

[22]  Bastian Goldlücke,et al.  Variational Analysis , 2014, Computer Vision, A Reference Guide.

[23]  P. Marcotte,et al.  A note on a globally convergent Newton method for solving monotone variational inequalities , 1986 .

[24]  Wenyu Sun,et al.  A modified projection method with a new direction for solving variational inequalities , 2009, Appl. Math. Comput..

[25]  Houyuan Jiang,et al.  A New Nonsmooth Equations Approach to Nonlinear Complementarity Problems , 1997 .

[26]  C. Perng On a class of theorems equivalent to Farkas's lemma , 2017 .

[27]  L. Liao,et al.  Improvements of Some Projection Methods for Monotone Nonlinear Variational Inequalities , 2002 .

[28]  Chao Zhang,et al.  A half thresholding projection algorithm for sparse solutions of LCPs , 2015, Optim. Lett..

[29]  Michael C. Ferris,et al.  Complementarity and variational problems : state of the art , 1997 .

[30]  Olvi L. Mangasarian,et al.  Nonlinear Programming , 1969 .

[31]  Christian Kanzow,et al.  On the resolution of monotone complementarity problems , 1996, Comput. Optim. Appl..

[32]  M. Ulbrich,et al.  A New Relaxation Scheme for Mathematical Programs with Equilibrium Constraints , 2010, SIAM J. Optim..

[33]  Francisco Facchinei,et al.  A semismooth equation approach to the solution of nonlinear complementarity problems , 1996, Math. Program..

[34]  Helmut Kleinmichel,et al.  A New Class of Semismooth Newton-Type Methods for Nonlinear Complementarity Problems , 1998, Comput. Optim. Appl..

[35]  A. Fischer A special newton-type optimization method , 1992 .

[36]  Leonard Yves Rüst The P-matrix linear complementarity problem , 2007 .

[37]  Jorge J. Moré,et al.  Global Methods for Nonlinear Complementarity Problems , 1994, Math. Oper. Res..

[38]  M. Fukushima,et al.  New NCP-Functions and Their Properties , 1997 .

[39]  Naihua Xiu,et al.  On Solutions of Sparsity Constrained Optimization , 2015 .

[40]  Patrick T. Harker,et al.  Smooth Approximations to Nonlinear Complementarity Problems , 1997, SIAM J. Optim..

[41]  Boris S. Mordukhovich,et al.  An Easy Path to Convex Analysis and Applications , 2013, Synthesis Lectures on Mathematics & Statistics.

[42]  C. Kanzow Nonlinear complementarity as unconstrained optimization , 1996 .

[43]  M. Fukushima Merit Functions for Variational Inequality and Complementarity Problems , 1996 .

[44]  F. Facchinei,et al.  Finite-Dimensional Variational Inequalities and Complementarity Problems , 2003 .