Computing the conjugate of convex piecewise linear-quadratic bivariate functions

We present a new algorithm to compute the Legendre–Fenchel conjugate of convex piecewise linear-quadratic (PLQ) bivariate functions. The algorithm stores a function using a (primal) planar arrangement. It then computes the (dual) arrangement associated with the conjugate by looping through vertices, edges, and faces in the primal arrangement and building associated dual vertices, edges, and faces. Using optimal computational geometry data structures, the algorithm has a linear time worst-case complexity. We present the algorithm, and illustrate it with numerical examples. We proceed to build a toolbox for convex bivariate PLQ functions by implementing the addition, and scalar multiplication operations. Finally, we compose these operators to compute classical convex analysis operators such as the Moreau envelope, and the proximal average.

[1]  Yves Lucet,et al.  Fast Moreau envelope computation I: numerical algorithms , 2007, Numerical Algorithms.

[2]  A. Noullez,et al.  A fast Legendre transform algorithm and applications to the adhesion model , 1994 .

[3]  J. Hiriart-Urruty,et al.  Convex analysis and minimization algorithms , 1993 .

[4]  Heinz H. Bauschke,et al.  Primal-Dual Symmetric Intrinsic Methods for Finding Antiderivatives of Cyclically Monotone Operators , 2007, SIAM J. Control. Optim..

[5]  Y. LUCET,et al.  A fast computational algorithm for the Legendre-Fenchel transform , 1996, Comput. Optim. Appl..

[6]  Yves Lucet,et al.  Faster than the Fast Legendre Transform, the Linear-time Legendre Transform , 1997, Numerical Algorithms.

[7]  Yves Lucet.  A linear Euclidean distance transform algorithm based on the linear-time Legendre transform , 2005, The 2nd Canadian Conference on Computer and Robot Vision (CRV'05).

[8]  Sarah Michelle Moffat.  On the kernel average for n functions , 2009 .

[9]  Heinz H. Bauschke,et al.  The Proximal Average: Basic Theory , 2008, SIAM J. Optim..

[10]  E. Aurell,et al.  The inviscid Burgers equation with initial data of Brownian type , 1992 .

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

[12]  Heinz H. Bauschke,et al.  Self-Dual Smooth Approximations of Convex Functions via the Proximal Average , 2011, Fixed-Point Algorithms for Inverse Problems in Science and Engineering.

[13]  Rafal Goebel,et al.  Self-dual smoothing of convex and saddle functions , 2007 .

[14]  Yves Lucet,et al.  Graph-Matrix Calculus for Computational Convex Analysis , 2011, Fixed-Point Algorithms for Inverse Problems in Science and Engineering.

[15]  L. Corrias.  Fast Legendre--Fenchel Transform and Applications to Hamilton--Jacobi Equations and Conservation Laws , 1996 .

[16]  Heinz H. Bauschke,et al.  Symbolic computation of Fenchel conjugates , 2006, ACCA.

[17]  Jie Sun.  On the structure of convex piecewise quadratic functions , 1992 .

[18]  Heinz H. Bauschke,et al.  The piecewise linear-quadratic model for computational convex analysis , 2009, Comput. Optim. Appl..

[19]  Michael T. Goodrich,et al.  Education forum: Web Enhanced Textbooks , 1998, SIGA.

[20]  Jonathan M. Borwein,et al.  Symbolic Fenchel Conjugation , 2008, Math. Program..

[21]  Yves Lucet,et al.  Convex Hull Algorithms for Piecewise Linear-Quadratic Functions in Computational Convex Analysis , 2010 .

[22]  J. Moreau.  Proximité et dualité dans un espace hilbertien , 1965 .

[23]  Y. Brenier.  Un algorithme rapide pour le calcul de transformées de Legendre-Fenchel discrètes , 1989 .

[24]  Yves Lucet,et al.  Convexity of the Proximal Average , 2011, J. Optim. Theory Appl..

[25]  George Havas,et al.  Perfect Hashing , 1997, Theor. Comput. Sci..

[26]  Yves Lucet,et al.  What Shape Is Your Conjugate? A Survey of Computational Convex Analysis and Its Applications , 2009, SIAM J. Optim..

[27]  Anísio Lacerda,et al.  Minimal perfect hashing: A competitive method for indexing internal memory , 2011, Inf. Sci..

[28]  David H. Bailey,et al.  Algorithms and applications , 1988 .

[29]  Heinz H. Bauschke,et al.  How to Transform One Convex Function Continuously into Another , 2008, SIAM Rev..

[30]  Heinz H. Bauschke,et al.  Projection and proximal point methods: convergence results and counterexamples , 2004 .