-
Computation of the Łojasiewicz exponents of real bivariate analytic functions
Authors:
Si Tiep Dinh,
Feng Guo,
Hong Duc Nguyen,
Tien Son Pham
Abstract:
The main goal of this paper is to present some explicit formulas for computing the {Ł}ojasiewicz exponent in the {Ł}ojasiewicz inequality comparing the rate of growth of two real bivariate analytic function germs.
The main goal of this paper is to present some explicit formulas for computing the {Ł}ojasiewicz exponent in the {Ł}ojasiewicz inequality comparing the rate of growth of two real bivariate analytic function germs.
△ Less
Submitted 10 May, 2024;
originally announced May 2024.
-
Existence theorems for optimal solutions in semi-algebraic optimization
Authors:
Jae Hyoung Lee,
Gue Myung Lee,
Tien Son Pham
Abstract:
Consider the problem of minimizing a lower semi-continuous semi-algebraic function $f \colon \mathbb{R}^n \to \mathbb{R} \cup \{+\infty\}$ on an unbounded closed semi-algebraic set $S \subset \mathbb{R}^n.$ Employing adequate tools of semi-algebraic geometry, we first establish some properties of the tangency variety of the restriction of $f$ on $S.$ Then we derive verifiable necessary and suffici…
▽ More
Consider the problem of minimizing a lower semi-continuous semi-algebraic function $f \colon \mathbb{R}^n \to \mathbb{R} \cup \{+\infty\}$ on an unbounded closed semi-algebraic set $S \subset \mathbb{R}^n.$ Employing adequate tools of semi-algebraic geometry, we first establish some properties of the tangency variety of the restriction of $f$ on $S.$ Then we derive verifiable necessary and sufficient conditions for the existence of optimal solutions of the problem as well as the boundedness from below and coercivity of the restriction of $f$ on $S.$ We also present a computable formula for the optimal value of the problem.
△ Less
Submitted 10 August, 2023;
originally announced August 2023.
-
Subdifferentials at infinity and applications in optimization
Authors:
Do Sang Kim,
Minh Tung Nguyen,
Tien Son Pham
Abstract:
In this work, the notions of normal cones at infinity to unbounded sets and limiting and singular subdifferentials at infinity for extended real value functions are introduced. Various calculus rules for these notions objects are established. A complete characterization of the Lipschitz continuity at infinity for lower semi-continuous functions is given. The obtained results are aimed ultimately a…
▽ More
In this work, the notions of normal cones at infinity to unbounded sets and limiting and singular subdifferentials at infinity for extended real value functions are introduced. Various calculus rules for these notions objects are established. A complete characterization of the Lipschitz continuity at infinity for lower semi-continuous functions is given. The obtained results are aimed ultimately at applications to diverse problems of optimization, such as optimality conditions, coercive properties, weak sharp minima and stability results.
△ Less
Submitted 28 July, 2023;
originally announced July 2023.
-
Characterizations of directional openness for set-valued map**s
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
We provide necessary and sufficient conditions for a set-valued map** between finite dimensional spaces to be directionally open by relating this property with directional regularity, Hölder continuity of the inverse map**, coderivatives and variations. These generalize and refine some previously known results.
We provide necessary and sufficient conditions for a set-valued map** between finite dimensional spaces to be directionally open by relating this property with directional regularity, Hölder continuity of the inverse map**, coderivatives and variations. These generalize and refine some previously known results.
△ Less
Submitted 25 July, 2022;
originally announced July 2022.
-
Limits of real bivariate rational functions
Authors:
Si Tiep Dinh,
Feng Guo,
Hong Duc Nguyen,
Tien Son Pham
Abstract:
Given two nonzero polynomials $f, g \in\mathbb R[x,y]$ and a point $(a, b) \in \mathbb{R}^2,$ we give some necessary and sufficient conditions for the existence of the limit $\displaystyle \lim_{(x, y) \to (a, b)} \frac{f(x, y)}{g(x, y)}.$ We also show that, if the denominator $g$ has an isolated zero at the given point $(a, b),$ then the set of possible limits of…
▽ More
Given two nonzero polynomials $f, g \in\mathbb R[x,y]$ and a point $(a, b) \in \mathbb{R}^2,$ we give some necessary and sufficient conditions for the existence of the limit $\displaystyle \lim_{(x, y) \to (a, b)} \frac{f(x, y)}{g(x, y)}.$ We also show that, if the denominator $g$ has an isolated zero at the given point $(a, b),$ then the set of possible limits of $\displaystyle \lim_{(x, y) \to (a, b)} \frac{f(x, y)}{g(x, y)}$ is a closed interval in $\overline{\mathbb{R}}$ and can be explicitly determined. As an application, we propose an effective algorithm to verify the existence of the limit and compute the limit (if it exists). Our approach is geometric and is based on Puiseux expansions.
△ Less
Submitted 10 February, 2022;
originally announced February 2022.
-
On definable open continuous map**s
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
For a definable continuous map** $f$ from a definable connected open subset $Ω$ of $\mathbb R^n$ into $\mathbb R^n,$ we show that the following statements are equivalent:
(i) The map** $f$ is open.
(ii) The fibers of $f$ are finite and the Jacobian of $f$ does not change sign on the set of points at which $f$ is differentiable.
(iii) The fibers of ${f}$ are finite and the set of points a…
▽ More
For a definable continuous map** $f$ from a definable connected open subset $Ω$ of $\mathbb R^n$ into $\mathbb R^n,$ we show that the following statements are equivalent:
(i) The map** $f$ is open.
(ii) The fibers of $f$ are finite and the Jacobian of $f$ does not change sign on the set of points at which $f$ is differentiable.
(iii) The fibers of ${f}$ are finite and the set of points at which $f$ is not a local homeomorphism has dimension at most $n - 2.$
As an application, we prove that Whyburn's conjecture is true for definable map**s: A definable open continuous map** of one closed ball into another which maps boundary homeomorphically onto boundary is necessarily a homeomorphism.
△ Less
Submitted 7 July, 2021; v1 submitted 3 June, 2021;
originally announced June 2021.
-
Some classical analysis results for continuous definable map**s
Authors:
Xuan Duc Ha Truong,
Tien Son Pham
Abstract:
In this paper, we show that some fundamental results for smooth map**s (e.g., the Brouwer degree formula, the implicit function and inverse function theorems, the mean value theorem, Sard's theorem, Hadamard's global invertibility criteria, Pourciau's surjectivity and openness results) have natural extensions for continuous map**s that are definable in o-minimal structures. The arguments rely…
▽ More
In this paper, we show that some fundamental results for smooth map**s (e.g., the Brouwer degree formula, the implicit function and inverse function theorems, the mean value theorem, Sard's theorem, Hadamard's global invertibility criteria, Pourciau's surjectivity and openness results) have natural extensions for continuous map**s that are definable in o-minimal structures. The arguments rely on nice properties of definable map**s and sets.
△ Less
Submitted 25 May, 2021;
originally announced May 2021.
-
Some variational properties of tangent directions at infinity of real algebraic sets
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
In this paper, we relate the set of asymptotic critical values of a polynomial function $f$ with the set of discontinuity of two functions, the multivalued function which associate to each value $t$ the set of tangent directions at infinity of the fiber $f^{-1}(t)$ and the composition of the $(n-2)$-dimensional volume function with the first one. This gives necessary conditions of equisingularity…
▽ More
In this paper, we relate the set of asymptotic critical values of a polynomial function $f$ with the set of discontinuity of two functions, the multivalued function which associate to each value $t$ the set of tangent directions at infinity of the fiber $f^{-1}(t)$ and the composition of the $(n-2)$-dimensional volume function with the first one. This gives necessary conditions of equisingularity at infinity for the family of the fibers of a real polynomial function.
△ Less
Submitted 21 May, 2021;
originally announced May 2021.
-
Nichtnegativstellensätze for definable functions in o-minimal structures
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
This paper addresses to Nichtnegativstellensätze for definable functions in o-minimal structures on $(\mathbb{R}, +, \cdot).$ Namely, let $f, g_1, \ldots, g_l \colon \mathbb{R}^n \to \mathbb{R}$ be definable $C^p$-functions ($p \ge 2$) and assume that $f$ is non-negative on $S := \{x \in \mathbb{R}^n \ | \ g_1(x) \ge 0, \ldots, g_l(x) \ge 0 \}.$ Under some natural hypotheses on zeros of $f$ in…
▽ More
This paper addresses to Nichtnegativstellensätze for definable functions in o-minimal structures on $(\mathbb{R}, +, \cdot).$ Namely, let $f, g_1, \ldots, g_l \colon \mathbb{R}^n \to \mathbb{R}$ be definable $C^p$-functions ($p \ge 2$) and assume that $f$ is non-negative on $S := \{x \in \mathbb{R}^n \ | \ g_1(x) \ge 0, \ldots, g_l(x) \ge 0 \}.$ Under some natural hypotheses on zeros of $f$ in $S,$ we show that $f$ is expressible in the form $f = φ_0 + \sum_{i = 1}^l φ_i g_i,$ where each $φ_i$ is a sum of squares of definable $C^{p - 2}$-functions. As a consequence, we derive global optimality conditions which generalize the Karush--Kuhn--Tucker optimality conditions for nonlinear optimization.
△ Less
Submitted 18 May, 2021;
originally announced May 2021.
-
The mountain pass theorem in terms of tangencies
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
This paper addresses the Mountain Pass Theorem for locally Lipschitz functions on finite-dimensional vector spaces in terms of tangencies. Namely, let $f \colon \mathbb R^n \to \mathbb R$ be a locally Lipschitz function with a mountain pass geometry. Let $$c := \inf_{γ\in \mathcal A}\max_{t\in[0,1]}f(γ(t)),$$ where $\mathcal{A}$ is the set of all continuous paths joining $x^*$ to $y^*.$ We show th…
▽ More
This paper addresses the Mountain Pass Theorem for locally Lipschitz functions on finite-dimensional vector spaces in terms of tangencies. Namely, let $f \colon \mathbb R^n \to \mathbb R$ be a locally Lipschitz function with a mountain pass geometry. Let $$c := \inf_{γ\in \mathcal A}\max_{t\in[0,1]}f(γ(t)),$$ where $\mathcal{A}$ is the set of all continuous paths joining $x^*$ to $y^*.$ We show that either $c$ is a critical value of $f$ or $c$ is a tangency value at infinity of $f.$ This reduces to the Mountain Pass Theorem of Ambrosetti and Rabinowitz in the case where the function $f$ is definable (such as, semi-algebraic) in an o-minimal structure.
△ Less
Submitted 15 May, 2021;
originally announced May 2021.
-
Stability of closedness of semi-algebraic sets under continuous semi-algebraic map**s
Authors:
Si Tiep Dinh,
Zbigniew Jelonek,
Tien Son Pham
Abstract:
Given a closed semi-algebraic set $X \subset \mathbb{R}^n$ and a continuous semi-algebraic map** $G \colon X \to \mathbb{R}^m,$ it will be shown that there exists an open dense semi-algebraic subset $\mathscr{U}$ of $L(\mathbb{R}^n, \mathbb{R}^m),$ the space of all linear map**s from $\mathbb{R}^n$ to $\mathbb{R}^m,$ such that for all $F \in \mathscr{U},$ the image $(F + G)(X)$ is a closed (se…
▽ More
Given a closed semi-algebraic set $X \subset \mathbb{R}^n$ and a continuous semi-algebraic map** $G \colon X \to \mathbb{R}^m,$ it will be shown that there exists an open dense semi-algebraic subset $\mathscr{U}$ of $L(\mathbb{R}^n, \mathbb{R}^m),$ the space of all linear map**s from $\mathbb{R}^n$ to $\mathbb{R}^m,$ such that for all $F \in \mathscr{U},$ the image $(F + G)(X)$ is a closed (semi-algebraic) set in $\mathbb{R}^m.$ To do this, we study the tangent cone at infinity $C_\infty X$ and the set $E_\infty X \subset C_\infty X$ of (unit) exceptional directions at infinity of $X.$ Specifically we show that the set $E_\infty X$ is nowhere dense in $C_\infty X \cap \mathbb{S}^{n - 1}.$
△ Less
Submitted 2 April, 2021; v1 submitted 10 October, 2020;
originally announced October 2020.
-
Limits of tangent spaces to definable sets
Authors:
Si Tiep Dinh,
Olivier Le Gal,
Tien Son Pham
Abstract:
We study the set of tangent limits at a given point to a set definable in any o-minimal structure by characterizing the set of exceptional rays in the tangent cone to the set at that point and investigating the set of tangent limits along these rays. Several criteria for determining exceptional rays will be given. The main results of the paper generalize, to the o-minimal setting and to arbitrary…
▽ More
We study the set of tangent limits at a given point to a set definable in any o-minimal structure by characterizing the set of exceptional rays in the tangent cone to the set at that point and investigating the set of tangent limits along these rays. Several criteria for determining exceptional rays will be given. The main results of the paper generalize, to the o-minimal setting and to arbitrary dimension, the main results of O'Shea--Wilson which deals with algebraic surfaces in $\mathbb R^3$.
△ Less
Submitted 7 October, 2020;
originally announced October 2020.
-
Stability of closedness of closed convex sets under linear map**s
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
We study the problem of when the continuous linear image of a fixed closed convex set $X \subset\mathbb{R}^n$ is closed. Specifically, we improve the main results in the papers \cite{Borwein2009, Borwein2010} by showing that for all, except for at most a $σ$-porous set, of the linear map**s $T$ from $\mathbb{R}^n$ into $\mathbb{R}^m,$ not only $T(X)$ is closed, but there is also an open neighbor…
▽ More
We study the problem of when the continuous linear image of a fixed closed convex set $X \subset\mathbb{R}^n$ is closed. Specifically, we improve the main results in the papers \cite{Borwein2009, Borwein2010} by showing that for all, except for at most a $σ$-porous set, of the linear map**s $T$ from $\mathbb{R}^n$ into $\mathbb{R}^m,$ not only $T(X)$ is closed, but there is also an open neighborhood of $T$ whose members also preserve the closedness of $X.$
△ Less
Submitted 2 April, 2021; v1 submitted 22 January, 2020;
originally announced January 2020.
-
Global mixed Łojasiewicz inequalities and asymptotic critical values
Authors:
Si Tiep Dinh,
Krzysztof Kurdyka,
Tien Son Pham
Abstract:
In this paper, we prove a version of global Łojasiewicz inequality for $C^1$ semialgebraic functions and relate its existence to the set of asymptotic critical values.
In this paper, we prove a version of global Łojasiewicz inequality for $C^1$ semialgebraic functions and relate its existence to the set of asymptotic critical values.
△ Less
Submitted 17 November, 2018;
originally announced November 2018.
-
Volume estimates of sublevel sets of real polynomials
Authors:
Nguyen Quang Dieu,
Dau Hoang Hung,
Tien Son Pham,
Hoang Thieu Anh
Abstract:
We give upper bounds for volume of sublevel sets of real polynomials. Our method is to combine a version of global Lojasiewicz inequality with some well known estimate on volume of tubes around real algebraic sets. Some applications to oscillatory integrals and integration indices of real polynomial are also given.
We give upper bounds for volume of sublevel sets of real polynomials. Our method is to combine a version of global Lojasiewicz inequality with some well known estimate on volume of tubes around real algebraic sets. Some applications to oscillatory integrals and integration indices of real polynomial are also given.
△ Less
Submitted 17 April, 2018; v1 submitted 13 November, 2017;
originally announced November 2017.
-
The bifurcation set of a real polynomial function of two variables and Newton polygons of singularities at infinity
Authors:
Masaharu Ishikawa,
Tat Thang Nguyen,
Tien Son Pham
Abstract:
In this paper, we determine the bifurcation set of a real polynomial function of two variables for non-degenerate case in the sense of Newton polygons by using a toric compactification. We also count the number of singular phenomena at infinity, called "cleaving" and "vanishing" in the same setting. Finally, we give an upper bound of the number of elements in the bifurcation set in terms of its Ne…
▽ More
In this paper, we determine the bifurcation set of a real polynomial function of two variables for non-degenerate case in the sense of Newton polygons by using a toric compactification. We also count the number of singular phenomena at infinity, called "cleaving" and "vanishing" in the same setting. Finally, we give an upper bound of the number of elements in the bifurcation set in terms of its Newton polygon. To obtain the upper bound, we apply toric modifications to the singularities at infinity successively.
△ Less
Submitted 8 August, 2016;
originally announced August 2016.
-
Łojasiewicz inequalities with explicit exponent for smallest singular value functions
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
Let $F(x) := (f_{ij}(x))_{i=1,\ldots,p; j=1,\ldots,q},$ be a ($p\times q$)-real polynomial matrix and let $f(x)$ be the smallest singular value function of $F(x).$ In this paper, we first give the following {\em nonsmooth} version of Łojasiewicz gradient inequality for the function $f$ with an explicit exponent: {\em For any $\bar x\in \Bbb R^n$, there exist $c > 0$ and $ε> 0$ such that we have fo…
▽ More
Let $F(x) := (f_{ij}(x))_{i=1,\ldots,p; j=1,\ldots,q},$ be a ($p\times q$)-real polynomial matrix and let $f(x)$ be the smallest singular value function of $F(x).$ In this paper, we first give the following {\em nonsmooth} version of Łojasiewicz gradient inequality for the function $f$ with an explicit exponent: {\em For any $\bar x\in \Bbb R^n$, there exist $c > 0$ and $ε> 0$ such that we have for all $\|x - \bar{x}\| < ε,$ \begin{equation*} \inf \{ \| w \| \ : \ w \in {\partial} f(x) \} \ \ge \ c\, |f(x)-f(\bar x)|^{1 - \frac{2}{\mathscr R(n+p,2d+2)}}, \end{equation*} where ${\partial} f(x)$ is the limiting subdifferential of $f$ at $x$, $d:=\max_{i=1,\ldots,p; j=1,\ldots,q}°f_{i j}$ and $\mathscr R(n, d) := d(3d - 3)^{n-1}$ if $d \ge 2$ and $\mathscr R(n, d) := 1$ if $d = 1.$} Then we establish some versions of Łojasiewicz inequality for the distance function with explicit exponents, locally and globally, for the smallest singular value function $f(x)$ of the matrix $F(x)$.
△ Less
Submitted 11 April, 2016;
originally announced April 2016.
-
Error Bounds for Parametric Polynomial Systems with Applications to Higher-Order Stability Analysis and Convergence Rates
Authors:
G. Li,
B. S. Mordukhovich,
T. T. A. Nghia,
T. S. Pham
Abstract:
The paper addresses parametric inequality systems described by polynomial functions in finite dimensions, where state-dependent infinite parameter sets are given by finitely many polynomial inequalities and equalities. Such systems can be viewed, in particular, as solution sets to problems of generalized semi-infinite programming with polynomial data. Exploiting the imposed polynomial structure to…
▽ More
The paper addresses parametric inequality systems described by polynomial functions in finite dimensions, where state-dependent infinite parameter sets are given by finitely many polynomial inequalities and equalities. Such systems can be viewed, in particular, as solution sets to problems of generalized semi-infinite programming with polynomial data. Exploiting the imposed polynomial structure together with powerful tools of variational analysis and semialgebraic geometry, we establish a far-going extension of the Łojasiewicz gradient inequality to the general nonsmooth class of supremum marginal functions as well as higher-order (Hölder type) local error bounds results with explicitly calculated exponents. The obtained results are applied to higher-order quantitative stability analysis for various classes of optimization problems including generalized semi-infinite programming with polynomial data, optimization of real polynomials under polynomial matrix inequality constraints, and polynomial second-order cone programming. Other applications provide explicit convergence rate estimates for the cyclic projection algorithm to find common points of convex sets described by matrix polynomial inequalities and for the asymptotic convergence of trajectories of subgradient dynamical systems in semialgebraic settings.
△ Less
Submitted 12 September, 2015;
originally announced September 2015.
-
Semidefinite approximations of the polynomial abscissa
Authors:
Roxana Heß,
Didier Henrion,
Jean-Bernard Lasserre,
Tien Son Pham
Abstract:
Given a univariate polynomial, its abscissa is the maximum real part of its roots. The abscissa arises naturally when controlling linear differential equations. As a function of the polynomial coefficients, the abscissa is H{ö}lder continuous, and not locally Lipschitz in general, which is a source of numerical difficulties for designing and optimizing control laws. In this paper we propose simple…
▽ More
Given a univariate polynomial, its abscissa is the maximum real part of its roots. The abscissa arises naturally when controlling linear differential equations. As a function of the polynomial coefficients, the abscissa is H{ö}lder continuous, and not locally Lipschitz in general, which is a source of numerical difficulties for designing and optimizing control laws. In this paper we propose simple approximations of the abscissa given by polynomials of fixed degree, and hence controlled complexity. Our approximations are computed by a hierarchy of finite-dimensional convex semidefinite programming problems. When their degree tends to infinity, the polynomial approximations converge in norm to the abcissa, either from above or from below.
△ Less
Submitted 30 July, 2015;
originally announced July 2015.
-
Convergent Semidefinite Programming Relaxations for Global Bilevel Polynomial Optimization Problems
Authors:
V. Jeyakumar,
J. B. Lasserre,
G. Li,
T. S. Pham
Abstract:
In this paper, we consider a bilevel polynomial optimization problem where the objective and the constraint functions of both the upper and the lower level problems are polynomials. We present methods for finding its global minimizers and global minimum using a sequence of semidefinite programming (SDP) relaxations and provide convergence results for the methods. Our scheme for problems with a con…
▽ More
In this paper, we consider a bilevel polynomial optimization problem where the objective and the constraint functions of both the upper and the lower level problems are polynomials. We present methods for finding its global minimizers and global minimum using a sequence of semidefinite programming (SDP) relaxations and provide convergence results for the methods. Our scheme for problems with a convex lower-level problem involves solving a transformed equivalent single-level problem by a sequence of SDP relaxations; whereas our approach for general problems involving a non-convex polynomial lower-level problem solves a sequence of approximation problems via another sequence of SDP relaxations.
△ Less
Submitted 13 January, 2016; v1 submitted 5 June, 2015;
originally announced June 2015.
-
Łojasiewicz-type inequalities with explicit exponents for the largest eigenvalue function of real symmetric polynomial matrices
Authors:
Si Tiep Dinh,
Tien Son Pham
Abstract:
Let $F(x) := (f_{ij}(x))_{i,j=1,\ldots,p},$ be a real symmetric polynomial matrix of order $p$ and let $f(x)$ be the largest eigenvalue function of the matrix $F(x).$ We denote by ${\partial}^\circ f(x)$ the Clarke subdifferential of $f$ at $x.$ In this paper, we first give the following {\em nonsmooth} version of Łojasiewicz gradient inequality for the function $f$ with an explicit exponent: For…
▽ More
Let $F(x) := (f_{ij}(x))_{i,j=1,\ldots,p},$ be a real symmetric polynomial matrix of order $p$ and let $f(x)$ be the largest eigenvalue function of the matrix $F(x).$ We denote by ${\partial}^\circ f(x)$ the Clarke subdifferential of $f$ at $x.$ In this paper, we first give the following {\em nonsmooth} version of Łojasiewicz gradient inequality for the function $f$ with an explicit exponent: For any $\bar x\in \Bbb R^n$ there exist $c > 0$ and $ε> 0$ such that we have for all $\|x - \bar{x}\| < ε,$ \begin{equation*} \inf \{ \| w \| \ : \ w \in {\partial}^\circ f(x) \} \ \ge \ c\, |f(x) - f(\bar x)|^{1 - \frac{1}{\mathscr{R}(2n+p(n+1),d+3)}}, \end{equation*} where $d:=\max_{i,j = 1, \ldots, p}°f_{i j}$ and $\mathscr{R}$ is a function introduced by D'Acunto and Kurdyka: $\mathscr{R}(n, d) := d(3d - 3)^{n-1}$ if $d \ge 2$ and $\mathscr{R}(n, d) := 1$ if $d = 1.$ Then we establish error bounds with explicitly determined exponents, local and global, for the largest eigenvalue function $f(x)$ of the matrix $F(x)$.
△ Less
Submitted 4 January, 2016; v1 submitted 7 January, 2015;
originally announced January 2015.
-
Convergence of the Lasserre Hierarchy of SDP Relaxations for Convex Polynomial Programs without Compactness
Authors:
V. Jeyakumar,
T. S. Pham,
G. Li
Abstract:
The Lasserre hierarchy of semidefinite programming (SDP) relaxations is an effective scheme for finding computationally feasible SDP approximations of polynomial optimization over compact semi-algebraic sets. In this paper, we show that, for convex polynomial optimization, the Lasserre hierarchy with a slightly extended quadratic module always converges asymptotically even in the face of non-compa…
▽ More
The Lasserre hierarchy of semidefinite programming (SDP) relaxations is an effective scheme for finding computationally feasible SDP approximations of polynomial optimization over compact semi-algebraic sets. In this paper, we show that, for convex polynomial optimization, the Lasserre hierarchy with a slightly extended quadratic module always converges asymptotically even in the face of non-compact semi-algebraic feasible sets. We do this by exploiting a coercivity property of convex polynomials that are bounded below. We further establish that the positive definiteness of the Hessian of the associated Lagrangian at a saddle-point (rather than the objective function at each minimizer) guarantees finite convergence of the hierarchy. We obtain finite convergence by first establishing a new sum-of-squares polynomial representation of convex polynomials over convex semi-algebraic sets under a saddle-point condition. We finally prove that the existence of a saddle-point of the Lagrangian for a convex polynomial program is also necessary for the hierarchy to have finite convergence.
△ Less
Submitted 27 June, 2013;
originally announced June 2013.