New Parallel Variants on Parallel Multi - Dimensional Search for Unconstrained Optimisation
نویسنده
چکیده
In this report, which continues the work by the same authors presented in Report 94.1 of the School of Computer Studies, we explore the effectiveness of three new variants on the method of Parallel Multi-Dimensional Search (PMDS) due to Torczon (1989). Since PMDS is acknowledged to be very slow at low dimensions we propose geometrically based variants which prove to be very effective over a range of test functions with dimensionality up to 32. We also identify some difficulties defined as 'restricted convergence' with PMDS which the new methods alleviate. Since the solution time for PMDS is reduced by factors of tens, hundreds and even thousands in some cases, this type of variant is clearly worth further research at higher dimensions and in adaptive forms. This research is also reported concisely in [2]
منابع مشابه
A Study of Distributed Evolutionary Algorithms for Multi-objective Optimisation
Most popular Evolutionary Algorithms for single multi-objective optimisation are motivated by the reduction of the computation time and the resolution larger problems. A promising alternative is to create new distributed schemes that improve the behaviour of the search process of such algorithms. In the multi-objective optimisation problems, more exploration of the search space is required to o...
متن کاملحل مسئله زمان بندی ماشینهای موازی نامرتبط با اهداف کل زودکرد وزنی و کل دیرکرد وزنی با استفاده از الگوریتم جستجوی پراکنده چند هدفه
The parallel machine scheduling problem is an important and difficult problem to be considered in the real-world situations. Traditionally, this problem consists of the scheduling of a set of independent jobs on parallel machines with the aim of minimizing the maximum job completion. In today's manufacturing systems, in which both early and tardy finishing of job processing are undesired, the o...
متن کاملA Modified Discreet Particle Swarm Optimization for a Multi-level Emergency Supplies Distribution Network
Currently, the research of emergency supplies distribution and decision models mostly focus on deterministic models and exact algorithm. A few of studies have been done on the multi-level distribution network and matheuristic algorithm. In this paper, random processes theory is adopted to establish emergency supplies distribution and decision model for multi-level network. By analyzing the char...
متن کاملA Hybrid Unconscious Search Algorithm for Mixed-model Assembly Line Balancing Problem with SDST, Parallel Workstation and Learning Effect
Due to the variety of products, simultaneous production of different models has an important role in production systems. Moreover, considering the realistic constraints in designing production lines attracted a lot of attentions in recent researches. Since the assembly line balancing problem is NP-hard, efficient methods are needed to solve this kind of problems. In this study, a new hybrid met...
متن کاملA new hybrid conjugate gradient algorithm for unconstrained optimization
In this paper, a new hybrid conjugate gradient algorithm is proposed for solving unconstrained optimization problems. This new method can generate sufficient descent directions unrelated to any line search. Moreover, the global convergence of the proposed method is proved under the Wolfe line search. Numerical experiments are also presented to show the efficiency of the proposed algorithm, espe...
متن کامل