Vaidya’s Method for Convex Stochastic Optimization Problems in Small Dimension

Mathematical Notes - The paper deals with a general problem of convex stochastic optimization in a space of small dimension (for example, 100 variables). It is known that for deterministic problems...     

Recursive Algorithms for Trailing Stop: Stochastic Approximation Approach

Trailing stops are often used in stock trading to limit the maximum of a possible loss and to lock in a profit. This work develops stochastic approximation algorithms to estimate the optimal trailing stop percentage. A stochastic optimization approach is proposed to recursively estimate the desired trailing stop percentage. A modification using projection is developed to ensure that the approximation sequence constructed stays in a reasonable range. Convergence of the algorithm is obtained. Moreover, interval estimates are constructed. Simulation examples are presented to compare our algorithm with Monte Carlo methods. Finally, we use real market data to demonstrate the algorithms.     

On Cone Characterizations of Strong and Lexicographic Optimality in Convex Multiobjective Optimization

Various type of optimal solutions of multiobjective optimization problems can be characterized by means of different cones. Provided the partial objectives are convex, we derive necessary and sufficient geometrical optimality conditions for strongly efficient and lexicographically optimal solutions by using the contingent, feasible and normal cones. Combining new results with previously known ones, we derive two general schemes reflecting the structural properties and the interconnections of five optimality principles: weak and proper Pareto optimality, efficiency and strong efficiency as well as lexicographic optimality.     

Convergence of Algorithms in Optimization and Solutions of Nonlinear Equations

It is desirable that an algorithm in unconstrained optimization converges when the guessed initial position is anywhere in a large region containing a minimum point. Furthermore, it is useful to have a measure of the rate of convergence which can easily be computed at every point along a trajectory to a minimum point. The Lyapunov function method provides a powerful tool to study convergence of iterative equations for computing a minimum point of a nonlinear unconstrained function or a solution of a system of nonlinear equations. It is surprising that this popular and powerful tool in the study of dynamical systems is not used directly to analyze the convergence properties of algorithms in optimization. We describe the Lyapunov function method and demonstrate how it can be used to study convergence of algorithms in optimization and in solutions of nonlinear equations. We develop an index which can measure the rate of convergence at all points along a trajectory to a minimum point and not just at points in a small neighborhood of a minimum point. Furthermore this index can be computed when the calculations are being carried out.     

Global Convergence of a Modified Gradient Projection Method for Convex Constrained Problems

In this paper, the continuously differentiable optimization problem min{f（x） ： x∈Ω}, where Ω ∈ R^n is a nonempty closed convex set, the gradient projection method by Calamai and More （Math. Programming, Vol.39. P.93-116, 1987） is modified by memory gradient to improve the convergence rate of the gradient projection method is considered. The convergence of the new method is analyzed without assuming that the iteration sequence {x^k} of bounded. Moreover, it is shown that, when f（x） is pseudo-convex （quasiconvex） function, this new method has strong convergence results. The numerical results show that the method in this paper is more effective than the gradient projection method.     

Beyond Convex? Global Optimization is Feasible Only for Convex Objective Functions: A Theorem

It is known that there are feasible algorithms for minimizing convex functions, and that for general functions, global minimization is a difficult (NP-hard) problem. It is reasonable to ask whether there exists a class of functions that is larger than the class of all convex functions for which we can still solve the corresponding minimization problems feasibly. In this paper, we prove, in essence, that no such more general class exists. In other words, we prove that global optimization is always feasible only for convex objective functions.     

The Million-Variable “March” for Stochastic Combinatorial Optimization

Combinatorial optimization problems have applications in a variety of sciences and engineering. In the presence of data uncertainty, these problems lead to stochastic combinatorial optimization problems which result in very large scale combinatorial optimization problems. In this paper, we report on the solution of some of the largest stochastic combinatorial optimization problems consisting of over a million binary variables. While the methodology is quite general, the specific application with which we conduct our experiments arises in stochastic server location problems. The main observation is that stochastic combinatorial optimization problems are comprised of loosely coupled subsystems. By taking advantage of the loosely coupled structure, we show that decomposition-coordination methods provide highly effective algorithms, and surpass the scalability of even the most efficiently implemented backtracking search algorithms.     

On Optimality Condition for Vector Optimization Based on Convex Set-Valued Mapping

1.IntroductionTheoptimalityconditionsofmathematicalprogrammingisaveryimportantsubjectbecausetLeyprovideausefulanalyticaltoolforstudingthedualitytheoryandnonlinearprogrammingalgoirthms.Inrecelltyears,someauthorshavebeguntostudytheoptimalityconditionsforvectoroptimizationproblemofset-valuedmapping,suchas[4][51.Inthispaperlwedefinetheconceptofcone--weaklyefficientsubdifferentialofset-valuedmappinginthecaseofgeneralpartiallyorderedlocallyconvextopologicalvectorspaces.Thecone-weaklysubdifferential…     

Switching Time Optimization for Bang-Bang and Singular Controls

Optimal control problems with the control variable appearing linearly are studied. A method for optimization with respect to the switching times of controls containing both bang-bang and singular arcs is presented. This method is based on the transformation of the control problem into a finite-dimensional optimization problem. Therein, first and second-order optimality conditions are thoroughly discussed. Explicit representations of first and second-order variational derivatives of the state trajectory with respect to the switching times are given. These formulas are used to prove that the second-order sufficient conditions can be verified on the basis of only first-order variational derivatives of the state trajectory. The effectiveness of the proposed method is tested with two numerical examples.     

The Projection onto a Direct Product of Convex Cones

ＴｈｅＰｒｏｊｅｃｔｉｏｎｏｎｔｏａＤｉｒｅｃｔｒｏｄｕｃｔｏｆＣｏｎｖｅｘＣｏｎｅｓＬｉｕＷｅｉ（刘维）ａｎｄＳｈｉＮｉｎｇｚｈｏｎｇ（史宁中）（ＤｅｐａｒｔｍｅｎｔｏｆＭａｔｈｅｍａｔｉｃｓ，ＮｏｒｔｈｅａｓｔＮｏｒｍａｌＵｎｉｖｅｒｓｉｔｙ，Ｃ...     

Hybrid Conjugate Gradient Algorithm for Unconstrained Optimization

In this paper a new hybrid conjugate gradient algorithm is proposed and analyzed. The parameter β _k is computed as a convex combination of the Polak-Ribière-Polyak and the Dai-Yuan conjugate gradient algorithms, i.e. β _k ^N =(1−θ _k )β _k ^PRP +θ _k β _k ^DY . The parameter θ _k in the convex combination is computed in such a way that the conjugacy condition is satisfied, independently of the line search. The line search uses the standard Wolfe conditions. The algorithm generates descent directions and when the iterates jam the directions satisfy the sufficient descent condition. Numerical comparisons with conjugate gradient algorithms using a set of 750 unconstrained optimization problems, some of them from the CUTE library, show that this hybrid computational scheme outperforms the known hybrid conjugate gradient algorithms. N. Andrei is a member of the Academy of Romanian Scientists, Splaiul Independenţei nr. 54, Sector 5, Bucharest, Romania.     

Weighted Poisson Cells as Models for Random Convex Polytopes

We introduce a parametric family for random convex polytopes in ? ^d which allows for an easy generation of samples for further use, e.g., as random particles in materials modelling and simulation. The basic idea consists in weighting the Poisson cell, which is the typical cell of the stationary and isotropic Poisson hyperplane tessellation, by suitable geometric characteristics. Since this approach results in an exponential family, parameters can be efficiently estimated by maximum likelihood. This work has been motivated by the desire for a flexible model for random convex particles as can be found in many composite materials such as concrete or refractory castables.     

Dual Representations for Convex Risk Measures via Conjugate Duality

The aim of this paper is to give dual representations for different convex risk measures by employing their conjugate functions. To establish the formulas for the conjugates, we use on the one hand some classical results from convex analysis and on the other hand some tools from the conjugate duality theory. Some characterizations of so-called deviation measures recently given in the literature turn out to be direct consequences of our results.     

Critical Landscape Topology for Optimization on the Symplectic Group

Optimization problems over compact Lie groups have been studied extensively due to their broad applications in linear programming and optimal control. This paper analyzes an optimization problem over a noncompact symplectic Lie group Sp(2N,ℝ), i.e., minimizing the Frobenius distance from a target symplectic transformation, which can be used to assess the fidelity function over dynamical transformations in classical mechanics and quantum optics. The topology of the set of critical points is proven to have a unique local minimum and a number of saddlepoint submanifolds, exhibiting the absence of local suboptima that may hinder the search for ultimate optimal solutions. Compared with those of previously studied problems on compact Lie groups, such as the orthogonal and unitary groups, the topology is more complicated due to the significant nonlinearity brought by the incompatibility of the Frobenius norm with the pseudo-Riemannian structure on the symplectic group.     

New Risk-Averse Control Paradigm for Stochastic Two-Time-Scale Systems and Performance Robustness

This work is concerned with the optimal control of stochastic two-time-scale linear systems with performance measure in a finite-horizon integral-quadratic form. Nature, modeled by stationary Wiener processes whose mean and covariance statistics are known, malevolently affects the state dynamics and output observations of the control problem class. With particular focus on the system performance robustness, the use of higher-order statistics or cumulants associated with the performance measure of chi-squared random variable type makes it possible to restate the stochastic control problem as the solution of a deterministic one, which subsequently allows disregarding all sample-path realizations by Nature acting on the original problem.     

Central Limit Theorems for a Super-Diffusion over a Stochastic Flow

Central limit theorems of the occupation time of a superprocess over a stochastic flow are proved. For the critical and higher dimensions d≥4, the limits are Gaussian variables. For d=3, the limit is conditional Gaussian. When the stochastic flow disappears, the results degenerate to those for the ordinary super-Brownian motion.     

New ODE Methods for Equality Constrained Optimization (2) — Algorithms

As a continuation of [1], this paper considers implementation of ODE approaches. A modified Hamming's algorithm for integration of (ECP)-equation is suggested to obtain a local solution. In addition to the main algorithm, three supporting algorithms are also described: two are for evaluation of the right-hand side of (ECP)-equation, which may be especially suitable for certain kinds of (ECP)-equation when applied to large scale problems; the third one, with a convergence theorem, is for computing an initial feasible point. Our numerical results obtained by executing these algorithms on an example of (ECP)-equation given in [1] on five test problems indicate their remarkable superiority of performance to Tanabe's ODE version that is recently claimed to be much better than some well-known SQP techniques.     

Auxiliary Principle and Iterative Algorithms for Lions-Stampacchia Variational Inequalities

In this paper, we extend the auxiliary principle (Cohen in J. Optim. Theory Appl. 49:325–333, 1988) to study a class of Lions-Stampacchia variational inequalities in Hilbert spaces. Our method consists in approximating, in the subproblems, the nonsmooth convex function by a sequence of piecewise linear and convex functions, as in the bundle method for nonsmooth optimization. This makes the subproblems more tractable. We show the existence of a solution for this Lions-Stampacchia variational inequality and explain how to build a new iterative scheme and a new stopping criterion. This iterative scheme and criterion are different from those commonly used in the special case of nonsmooth optimization. We study also the convergence of iterative sequences generated by the algorithm. This work was supported by the National Natural Science Foundation of China (10671135), the Specialized Research Fund for the Doctoral Program of Higher Education (20060610005), the National Natural Science Foundation of Sichuan Education Department of China (07ZB068) and the Open Fund (PLN0703) of State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation (Southwest Petroleum University).