Search results for "optimization"
showing 10 items of 2824 documents
Kernelizing LSPE(λ)
2007
We propose the use of kernel-based methods as underlying function approximator in the least-squares based policy evaluation framework of LSPE(λ) and LSTD(λ). In particular we present the 'kernelization' of model-free LSPE(λ). The 'kernelization' is computationally made possible by using the subset of regressors approximation, which approximates the kernel using a vastly reduced number of basis functions. The core of our proposed solution is an efficient recursive implementation with automatic supervised selection of the relevant basis functions. The LSPE method is well-suited for optimistic policy iteration and can thus be used in the context of online reinforcement learning. We use the hig…
Achieving Fair Load Balancing by Invoking a Learning Automata-Based Two-Time-Scale Separation Paradigm.
2020
Author's accepted manuscript. © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. In this article, we consider the problem of load balancing (LB), but, unlike the approaches that have been proposed earlier, we attempt to resolve the problem in a fair manner (or rather, it would probably be more appropriate to describe it as an ε-fair manner because, although the LB…
The design of absorbing Bayesian pursuit algorithms and the formal analyses of their ε-optimality
2016
The fundamental phenomenon that has been used to enhance the convergence speed of learning automata (LA) is that of incorporating the running maximum likelihood (ML) estimates of the action reward probabilities into the probability updating rules for selecting the actions. The frontiers of this field have been recently expanded by replacing the ML estimates with their corresponding Bayesian counterparts that incorporate the properties of the conjugate priors. These constitute the Bayesian pursuit algorithm (BPA), and the discretized Bayesian pursuit algorithm. Although these algorithms have been designed and efficiently implemented, and are, arguably, the fastest and most accurate LA report…
Hydropower Optimization Using Split-Window, Meta-Heuristic and Genetic Algorithms
2019
In this paper, we try to find the most efficient optimization algorithm that can be used to resolve the hydropower optimization problem. We propose a novel optimization technique is called the Split-window method. The method is relatively simple and reduces the complexity of the optimization problem by split-ting the planning horizon (and datasets) into equal windows and assigning the same values to policies(actions) within each part. After splitting, a meta-heuristic technique is used to optimize the actions, and the dataset is split again until a split contains only one instance (timestep). The unique values to be optimized during each iteration is equal to the number of splits which make…
Robust control of uncertain multi-inventory systems via linear matrix inequality
2008
We consider a continuous time linear multi inventory system with unknown demands bounded within ellipsoids and controls bounded within ellipsoids or polytopes. We address the problem of "-stabilizing the inventory since this implies some reduction of the inventory costs. The main results are certain conditions under which "-stabilizability is possible through a saturated linear state feedback control. All the results are based on a Linear Matrix Inequalities (LMIs) approach and on some recent techniques for the modeling and analysis of polytopic systems with saturations.
Optimal placement of 3D sensors considering range and field of view
2017
This paper describes a novel approach to the problem of optimal placement of 3D sensors in a specified volume of interest. The coverage area of the sensors is modelled as a cone having limited field of view and range. The volume of interest is divided into many, smaller cubes each having a set of associated Boolean and continuous variables. The proposed method could be easily extended to handle the case where certain sub-volumes must be covered by several sensors (redundancy), for example ex-zones, regions where humans are not allowed to enter or regions where machine movement may obstruct the view of a single sensor. The optimisation problem is formulated as a Mixed-Integer Linear Program …
A fuzzy method to repair infeasibility in linearly constrained problems
2001
Abstract In this paper we introduce a fuzzy method to deal with infeasibility in linearly constrained programs. Given an infeasible instance, we determine how much we should perturb the right-hand side coefficients in order to attain feasibility and propose a ‘feasible reformulation’ of the problem. Although we prove that our algorithm always finds such a reformulation the convenience of using it can be decided by the analyst. By this, we mean that the method also provides a simple way to compute lower bounds on the changes on every right-hand side coefficient, and if the decision maker considers that some of the magnitudes are unacceptable, he or she simply stops at this step. We think tha…
Mathematical Programming Methods for the Evaluation of Dynamic Plastic Deformations
1990
Dynamic plastic deformation can be evaluated with two accuracy levels, nemely either by a full analysis making use of a step-by-step procedure, or by a simplified analysis making use of a bounding technique. Both procedures can be achieved by means a unified mathematical programming approach here presented. It is shown that for a full analysis both the direct and indirect methods of linear dynamics coupled with mathematical programming methods can be successfully applied, whereas for a simplified analysis a convergent bounding principle, holding both below and above the shakedown limit, can be utilized to produce an efficient linear programming-based algorithm.
A choice of bilevel linear programming solving parameters: factoraggregation approach
2013
Our paper deals with the problem of choosing correct parameters for the bilevel linear program- ming solving algorithm proposed by M. Sakawa and I. Nishizaki. We suggest an approach based on fac- toraggregation, which is a specially designed general aggregation operator. The idea of factoraggregation arises from factorization by the equivalence relation generated by the upper level objective function. We prove several important properties of the factorag- gregation result regarding the analysis of param- eters in order to find an optimal solution for the problem. We illustrate the proposed method with some numerical and graphical examples, in particu- lar we consider a modification of the m…
Optimization under Uncertainty and Linear Semi-Infinite Programming: A Survey
2001
This paper deals with the relationship between semi-infinite linear programming and decision making under uncertainty in imprecise environments. Actually, we have reviewed several set-inclusive constrained models and some fuzzy programming problems in order to see if they can be solved by means of a linear semi-infinite program. Finally, we present some numerical examples obtained by using a primal semi-infinite programming method.