Approximate Analytic Center Research Articles

A NONLINEAR FEASIBILITY PROBLEM HEURISTIC

In this work we consider a region S ⊂ given by a finite number of nonlinear smooth convex inequalities and having nonempty interior. We assume a point x 0 is given, which is close in certain norm to the analytic center of S, and that a new nonlinear smooth convex inequality is added to those defining S (perturbed region). It is constructively shown how to obtain a shift of the right-hand side of this inequality such that the point x 0 is still close (in the same norm) to the analytic center of this shifted region. Starting from this point and using the theoretical results shown, we develop a heuristic that allows us to obtain the approximate analytic center of the perturbed region. Then, we present a procedure to solve the problem of nonlinear feasibility. The procedure was implemented and we performed some numerical tests for the quadratic (random) case.

An Analytic Center Cutting Plane Approach for Conic Programming

We analyze the problem of finding a point strictly interior to a bounded, convex, and fully dimensional set from a finite dimensional Hilbert space. We generalize the results obtained for the linear programming (LP), semidefinite programming (SDP), and second-order core programming (SOCP) cases. The cuts added by our algorithm are central and conic. In our analysis, we find an upper bound for the number of Newton steps required to compute an approximate analytic center. Also, we provide an upper bound for the total number of cuts added to solve the problem. This bound depends on the quality of the cuts, the dimensionality of the problem and the thickness of the set we are considering.

A second-order cone cutting surface method: complexity and application

We present an analytic center cutting surface algorithm that uses mixed linear and multiple second-order cone cuts. Theoretical issues and applications of this technique are discussed. From the theoretical viewpoint, we derive two complexity results. We show that an approximate analytic center can be recovered after simultaneously adding p second-order cone cuts in O(plog?(p+1)) Newton steps, and that the overall algorithm is polynomial. From the application viewpoint, we implement our algorithm on mixed linear-quadratic-semidefinite programming problems with bounded feasible region and report some computational results on randomly generated fully dense problems. We compare our CPU time with that of SDPLR, SDPT3, and SeDuMi and show that our algorithm outperforms these software packages on problems with fully dense coefficient matrices. We also show the performance of our algorithm on semidefinite relaxations of the maxcut and Lovasz theta problems.

ACCPM with a nonlinear constraint and an active set strategy to solve nonlinear multicommodity flow problems

This paper proposes an implementation of a constrained analytic center cutting plane method to solve nonlinear multicommodity flow problems. The new approach exploits the property that the objective of the Lagrangian dual problem has a smooth component with second order derivatives readily available in closed form. The cutting planes issued from the nonsmooth component and the epigraph set of the smooth component form a localization set that is endowed with a self-concordant augmented barrier. Our implementation uses an approximate analytic center associated with that barrier to query the oracle of the nonsmooth component. The paper also proposes an approximation scheme for the original objective. An active set strategy can be applied to the transformed problem: it reduces the dimension of the dual space and accelerates computations. The new approach solves huge instances with high accuracy. The method is compared to alternative approaches proposed in the literature.

An Interior Point Cutting Plane Method for the Convex Feasibility Problem with Second-Order Cone Inequalities

The convex feasibility problem in general is a problem of finding a point in a convex set that contains a full dimensional ball and is contained in a compact convex set. We assume that the outer set is described by second-order cone inequalities and propose an analytic center cutting plane technique to solve this problem. We discuss primal and dual settings simultaneously. Two complexity results are reported: the complexity of restoration procedure and complexity of the overall algorithm. We prove that an approximate analytic center is updated after adding a second-order cone cut (SOCC) in O(1) Newton step, and that the analytic center cutting plane method (ACCPM) with SOCC is a fully polynomial algorithm.

Analytic Center Cutting-Plane Method with Deep Cuts for Semidefinite Feasibility Problems

An analytic center cutting-plane method with deep cuts for semidefinite feasibility problems is presented. Our objective in these problems is to find a point in a nonempty bounded convex set Γ in the cone of symmetric positive-semidefinite matrices. The cutting plane method achieves this by the following iterative scheme. At each iteration, a query point Ŷ that is an approximate analytic center of the current working set is chosen. We assume that there exists an oracle which either confirms that Ŷ ∈Γ or returns a cut A ∈S m {Y∈S m : A●Y≤ A●YŶ - ξ} ⊃ Γ, where ξ ≥ 0. If Ŷ ∈Γ, an approximate analytic center of the new working set, defined by adding the new cut to the preceding working set, is then computed via a primal Newton procedure. Assuming that Γ contains a ball with radius ∈ > 0, the algorithm obtains eventually a point in Γ, with a worst-case complexity of O *(m 3/∈2) on the total number of cuts generated.

An Analytic Center Cutting Plane Method for Semidefinite Feasibility Problems

Semidefinite feasibility problems arise in many areas of operations research. The abstract form of these problems can be described as finding a point in a nonempty bounded convex body Γ in the cone of symmetric positive semidefinite matrices. Assume that Γ is defined by an oracle, which for any given m × m symmetric positive semidefinite matrix Ŷ either confirms that Ŷ ∈ Γ or returns a cut, i.e., a symmetric matrix A such that Γ is in the half-space {Y : A · Y ≤ A · Ŷ}. We study an analytic center cutting plane algorithm for this problem. At each iteration, the algorithm computes an approximate analytic center of a working set defined by the cutting plane system generated in the previous iterations. If this approximate analytic center is a solution, then the algorithm terminates; otherwise the new cutting plane returned by the oracle is added into the system. As the number of iterations increases, the working set shrinks and the algorithm eventually finds a solution to the problem. All iterates generated by the algorithm are positive definite matrices. The algorithm has a worst-case complexity of O*(m3/ε2) on the total number of cuts to be used, where ε is the maximum radius of a ball contained by Γ.

A Polynomial Cutting Surfaces Algorithm for the ConvexFeasibility Problem Defined by Self-Concordant Inequalities

Consider a nonempty convex set in R^m which is defined by a finite number of smooth convex inequalities and which admits a self-concordant logarithmic barrier. We study the analytic center based column generation algorithm for the problem of finding a feasible point in this set. At each iteration the algorithm computes an approximate analytic center of the set defined by the inequalities generated in the previous iterations. If this approximate analytic center is a solution, then the algorithm terminatess otherwise either an existing inequality is shifted or a new inequality is added into the system. As the number of iterations increases, the set defined by the generated inequalities shrinks and the algorithm eventually finds a solution of the problem. The algorithm can be thought of as an extension of the classical cutting plane method. The difference is that we use analytic centers and “convex cuts” instead of arbitrary infeasible points and linear cuts. In contrast to the cutting plane method, the algorithm has a polynomial worst case complexity ofO(N\log\frac{1}{\varepsilon}) on the total number of cuts to be used, where N is the number of convex inequalities in the original problem and e is the maximum common slack of the original inequality system.

Homogeneous analytic center cutting plane methods with approximate centers

In this paper we consider a homogeneous analytic center cutting plane method in a projective space. We describe a general scheme that uses a homogeneous oracle and computes an approximate analytic center at each iteration. This technique is applied to a convex feasibility problem, to variational inequalities, and to convex constrained minimization. We prove that these problems can be solved with the same order of complexity as in the case of exact analytic centers. For the feasibility and the minimization problems rough approximations suffice, but very high precision is required for the variational inequalities. We give an example of variational inequality where even the first analytic center needs to be computed with a precision matching the precision required for the solution.

A Nonlinear Analytic Center Cutting Plane Method for a Class of Convex Programming Problems

A cutting plane algorithm for minimizing a convex function subject to constraints defined by a separation oracle is presented. The algorithm is based on approximate analytic centers. The nonlinearity of the objective function is taken into account, yet the feasible region is approximated by a containing polytope. This containing polytope is regularly updated by adding a new cut through a test point. Each test point is an approximate analytic center of the intersection of a containing polytope and a level set of the nonlinear objective function. We establish the complexity of the algorithm. Our complexity estimate is given in terms of the problem dimension, the desired accuracy of an approximate solution, and other parameters that depend on the geometry of a specific instance of the problem.

An Analytic Center Based Column Generation Algorithm for Convex Quadratic Feasibility Problems

We consider the analytic center based column generation algorithm for the problem of finding a feasible point in a set defined by a finite number of convex quadratic inequalities. At each iteration the algorithm computes an approximate analytic center of the set defined by the intersection of quadratic inequalities generated in the previous iterations. If this approximate analytic center is a solution, then the algorithm terminates; otherwise a quadratic inequality violated at the current center is selected and a new quadratic cut (defined by a convex quadratic inequality) is placednear the approximate center. As the number of cuts increases, the set defined by their intersection shrinks and the algorithm eventually finds a solution to the problem. Previously, similar analytic center based column generation algorithms were studied first for the linear feasibility problem and later for the general convex feasibility problem. Our method differs from these early methods in that we use "quadratic cuts" in the computation instead of linear cuts. Moreover, our method has a polynomial worst case complexity of $O(n\ln \frac{1}{\varepsilon})$ on the total number of cuts to be used, where n is the number of convex quadratic polynomial inequalities in the problem and $\varepsilon$ is the radius of the largest ball contained in the feasible set. In contrast, the early column generation methods using linear cuts can only solve the convex quadratic feasibility problem in pseudopolynomial time.

Analysis of a Cutting Plane Method That Uses Weighted Analytic Center and Multiple Cuts

We consider the analytic center cutting plane (or column generation) algorithm for solving general convex problems defined by a separation oracle. The oracle is called at an approximate analytic center of a polytope which contains the solution set and is given by the intersection of the linear inequalities previously generated from the oracle. If the approximate center is not in the solution set, separating hyperplanes will be placed through the approximate center, and a new approximate analytic center will be found for the shrunken polytope. In this paper, we consider using approximate weighted analytic centers in the cutting plane method and show that the method, with multiple cuts added in each step, has a complexity of $O(\eta m^2/\epsilon^2)$, where $\eta$ is the maximum number of cuts that can be added in each step and $m$ is the dimension of the problem.

A note on some analytic center cutting plane methods for convex feasibility and minimization problems

Recently Goffin, Luo and Ye have analyzed the complexity of an analytic center algorithm for convex feasibility problems defined by a separation oracle. The oracle is called at the (possibly approximate) analytic center of the set given by the linear inequalities which are the previous answers of the oracle. We discuss oracles for the problem of minimizing a convex (possibly nondifferentiable) function subject to box constraints, and give corresponding complexity estimates.

Decomposition and Nondifferentiable Optimization with the Projective Algorithm

This paper deals with an application of a variant of Karmarkar's projective algorithm for linear programming to the solution of a generic nondifferentiable minimization problem. This problem is closely related to the Dantzig-Wolfe decomposition technique used in large-scale convex programming. The proposed method is based on a column generation technique defining a sequence of primal linear programming maximization problems. Associated with each problem one defines a weighted potential function which is minimized using a variant of the projective algorithm. When a point close to the minimum of the potential function is reached, a corresponding point in the dual space is constructed, which is close to the analytic center of a polytope containing the solution set of the nondifferentiable optimization problem. An admissible cut of the polytope, corresponding to a new supporting hyperplane of the epigraph of the function to minimize, is then generated at this approximate analytic center. In the primal space this new cut translates into a new column for the associated linear programming problem. The algorithm has performed well on a set of convex nondifferentiable programming problems.

An Algorithm for Convex Quadratic Programming That Requires O(n3.5L) Arithmetic Operations

A new interior point method for minimizing a convex quadratic function over a polytope is developed. We show that our method requires O(n3.5L) arithmetic operations. In our algorithm we construct a sequence Pz0, Pz1, …, Pzk, … of nested convex sets that shrink towards the set of optimal solution(s). During iteration k we take a partial Newton step to move from an approximate analytic center of Pzk − 1 to an approximate analytic center of Pzk. A system of linear equations is solved at each iteration to find the step direction. The solution that is available after O(√mL) iterations can be converted to an optimal solution. Our analysis indicates that inexact solutions to the linear system of equations could be used in implementing this algorithm.