Designing algorithms for an optimization model often amounts to maintaining a balance between the degree of information to request from the model on the one hand, and the computational speed to expect on the other hand. Naturally, the more information is available, the faster one can expect the algorithm to converge. The popular algorithm of … Read more
We propose a limited memory steepest descent (LMSD) method for solving unconstrained optimization problems. As a steepest descent method, the step computation in each iteration requires the evaluation of a gradient of the objective function and the calculation of a scalar step size only. When employed to solve certain convex problems, our method reduces to … Read more
We address the inverse problem of Lagrangian identification based on trajectories in the context of nonlinear optimal control. We propose a general formulation of the inverse problem based on occupation measures and complementarity in linear programming. The use of occupation measures in this context offers several advantages from the theoretical, numerical and statistical points of … Read more
We propose a dynamic program to find the shortest path in a network having exponential, gamma and normal probability distributions as arc lengths. Two operators of sum and comparison need to be adapted for the proposed dynamic program. Convolution approach is used to sum probability distributions being employed in the dynamic program. ArticleDownload View PDF
In this work, we study directional versions of the H\”olderian/Lipschitzian metric subregularity of multifunctions. Firstly, we establish variational characterizations of the H\”olderian/Lipschitzian directional metric subregularity by means of the strong slopes and next of mixed tangency-coderivative objects . By product, we give second-order conditions for the directional Lipschitzian metric subregularity and for the directional metric … Read more
Bottom-up search methods for determining the efficient set of a multiple objective linear programming (MOLP) problem have a valuable advantage that they can quickly give efficient subsets of the MOLP problem to the decision makers. Main difficulties of the previously appeared bottom-up search methods are finding all efficient extreme points adjacent to and enumerating all … Read more
In this paper, we present a new scheme of a sampling method to solve chance constrained programs. First of all, a modified sample average approximation, namely Partial Sample Average Approximation (PSAA) is presented. The main advantage of our approach is that the PSAA problem has only continuous variables whilst the standard sample average approximation (SAA) … Read more
We consider totally unimodular multistage stochastic programs, that is, multistage stochastic programs whose extensive-form constraint matrices are totally unimodular. We establish several sufficient conditions and identify examples that have arisen in the literature. CitationRuichen (Richard) Sun, Oleg V. Shylo, Andrew J. Schaefer, Totally unimodular multistage stochastic programs, Operations Research Letters, Volume 43, Issue 1, January … Read more
The multi-hour bandwidth packing problem arises in telecommunication networks that span several time horizon. The problem seeks to select and route a set of messages from a given list of messages with prespecified requirement on demand for bandwidth under time varying traffic conditions on an undirected communication network such that the total profit is maximized. … Read more
The portfolio optimization model has limited impact in practice due to estimation issues when applied with real data. To address this, we adapt two machine learning methods, regularization and cross-validation, for portfolio optimization. First, we introduce performance-based regularization (PBR), where the idea is to constrain the sample variances of the estimated portfolio risk and return, … Read more