An Improved Spsa Algorithm for Stochastic Optimization with Bound Constraints
نویسندگان
چکیده
We show that the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm with projection may exhibit slow convergence in constrained stochastic optimization problems when the optimum is situated on the constraints. The cause of the slow convergence is a geometric interaction between the projection operator and the SPSA gradient estimate. The effect of this interaction can be described as “bouncing of iterates against the constraints.” We describe this on two low dimensional noise-free examples, and present a new algorithm that does not exhibit the bouncing effect and the consequent slow convergence. Copyright c ©2005 IFAC
منابع مشابه
An Algorithm Based on Theory of Constraints and Branch and Bound for Solving Integrated Product-Mix-Outsourcing Problem
One of the most important decision making problems in many production systems is identification and determination of products and their quantities according to available resources. This problem is called product-mix. However, in the real-world situations, for existing constrained resources, many companies try to provide some products from external resources to achieve more profits. In this pape...
متن کاملA New Hybrid Simulated Annealing Algorithm for Large Scale Global Optimization
The present paper focus on the improvement of the efficiency of structural optimization, in typical structural optimization problems there may be many locally minimum configurations. For that reason, the application of a global method, which may escape from the locally minimum points, remain essential. In this paper, a new hybrid simulated annealing algorithm for large scale global optimization...
متن کامل2001: Constrained Optimization over Discrete Sets via Spsa with Application to Non-separable Resource Allocation
This paper presents a version of the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm for optimizing non-separable functions over discrete sets under given constraints. The primary motivation for discrete SPSA is to solve a class of resource allocation problems wherein the goal is to distribute a finite number of discrete resources to finitely many users in such a way as to o...
متن کاملRevisiting the Application of Simultaneous Perturbation Stochastic Approximation towards Signal Timing Optimization
Simultaneous Perturbation Stochastic Approximation (SPSA) has gained favor as an efficient optimization method for calibrating computationally intensive, “black box” traffic flow simulations. Few recent studies have investigated the efficiency of SPSA for traffic signal timing optimization. It is important for this to be investigated, because significant room for improvement exists in the area ...
متن کاملAUTOMATED SIZING OF TRUSS STRUCTURES USING A COMPUTATIONALLY IMPROVED SOPT ALGORITHM
The present study attempts to apply an efficient yet simple optimization (SOPT) algorithm to optimum design of truss structures under stress and displacement constraints. The computational efficiency of the technique is improved through avoiding unnecessary analyses during the course of optimization using the so-called upper bound strategy (UBS). The efficiency of the UBS integrated SOPT algori...
متن کامل