In this paper, we review the population learning algorithm and discuss its convergence. This algorithm was proposed as a tool for solving optimisation problems, and the concept underlying this approach is embedded in ...
详细信息
In this paper, we review the population learning algorithm and discuss its convergence. This algorithm was proposed as a tool for solving optimisation problems, and the concept underlying this approach is embedded in social educational processes. A convergence analysis of the algorithm is presented by means of a finite Markov chain analysis and by comparing its behaviour to evolutionary strategies in a process of searching for a global solution in a finite number of stages. The proposed populationalgorithm is also shown to be an alternative tool for solving different optimisation problems.
Radial Basis Function Neural Networks (RBFNs) are nowadays quite popular due to their ability to discover and approximate complex nonlinear dependencies within the data under analysis. Performance of the RBF network d...
详细信息
Radial Basis Function Neural Networks (RBFNs) are nowadays quite popular due to their ability to discover and approximate complex nonlinear dependencies within the data under analysis. Performance of the RBF network depends on numerous factors related to its initialization and training. The paper proposes an approach to the radial basis function networks design, where initial parameters of the network, output weights and parameters of the transfer function are set using the proposed agent-based population learning algorithm (PLA). The algorithm is validated experimentally. Advantages and main features of the PLA-based RBF designs are discussed basing on results of the computational experiment.
population-based hybrid metaheuristics, often inspired by biological or social phenomena, belong to a widely used groups of methods suitable for solving complex hard optimization problems. Their effectiveness has been...
详细信息
population-based hybrid metaheuristics, often inspired by biological or social phenomena, belong to a widely used groups of methods suitable for solving complex hard optimization problems. Their effectiveness has been confirmed for providing good quality solutions to many real-life instances of different problems. Recently, an incorporation of the cooperative problem solving paradigm into metaheuristics has become an interesting extension of the population-based hybrid metaheuristics. Cooperation is meant as a problem-solving strategy, consisting of a search performed by different search agents running in parallel. During the search, the agents cooperate by exchanging information about states, solutions or other search space characteristics. This paper proposes an Agent-Based Cooperative population learning algorithm for the Vehicle Routing Problem with Time Windows. In the proposed approach the process of search for the best solution is divided into stages, and different search procedures are used at each stage. These procedures use a set of various heuristics (represented by software agents) which run under the cooperation scheme defined separately for each stage. Computational experiment which has been carried out, confirmed the effectiveness of the proposed approach. (C) 2014 Elsevier B.V. All rights reserved.
The multi-mode resource-constrained project scheduling problem with minimum and maximum time lags is considered in the paper. An activity is performed in a mode, which determines the demand of renewable and nonrenewab...
详细信息
ISBN:
(纸本)9783642166921
The multi-mode resource-constrained project scheduling problem with minimum and maximum time lags is considered in the paper. An activity is performed in a mode, which determines the demand of renewable and nonrenewable resources required for its processing and minimum and maximum time lags between adjacent activities. The goal is to find a mode assignment to the activities and their start times such that all constraints are satisfied and the project duration is minimized. Because the problem is NP-hard a population-learningalgorithm (PLA2) is proposed to tackle the problem. PLA2 is a population-based approach which takes advantage of the features common to the social education system rather than to the evolutionary processes. The proposed approach perfectly suits for multi-agent systems because it is based on the idea of constructing a hybrid algorithm integrating different optimization techniques complementing each other and producing a synergetic effect. Results of the experiment were compared to the results published in Project Scheduling Problem Library.
The aim of the paper is to propose and evaluate an agent-based population learning algorithm generating, by the prototype selection, a representative training clataset of the required size. It is assumed that prototyp...
详细信息
ISBN:
(纸本)9783642239373;9783642239380
The aim of the paper is to propose and evaluate an agent-based population learning algorithm generating, by the prototype selection, a representative training clataset of the required size. It is assumed that prototypes are selected from clusters. The process of selection is executed by a team of agents, which execute various local search procedures and cooperate to find-out a solution to the considered problem. Rules for agent cooperation are defined within working strategies. In this paper influence of two different strategies and the population size on performance of the algorithm is investigated.
In the paper, we consider a population learning algorithm denoted (PLA3), with the differential evolution method for solving the discrete-continuous scheduling problem (DCSP) with continuous resource discretisation - ...
详细信息
ISBN:
(纸本)9781467364690
In the paper, we consider a population learning algorithm denoted (PLA3), with the differential evolution method for solving the discrete-continuous scheduling problem (DCSP) with continuous resource discretisation - Theta(Z). The considered problem originates from DCSP, in which nonpreemtable tasks should be scheduled on parallel identical machines under constraint on discrete resource and requiring, additionally, a renewable continuous resource to minimize the schedule length. The continuous resource in DCSP is divisible continuously and is allocated to tasks from a given interval in amounts unknown in advance. Task processing rate depends on the allocated amount of the continuous resource. To eliminate time consuming optimal continuous resource allocation, an NP-hard problem Theta(Z) with continuous resource discretisation is introduced and sub-optimally solved by PLA3. Experimental results show that PLA3 was able to improve best-known solutions and excels its predecessor PLA2 in solving the considered problem.
The paper investigates a possibility of combining the population learning algorithm and the A-Team concept with a view to increase quality of results and efficiency of computations. To implement the idea a middleware ...
详细信息
ISBN:
(纸本)9783540748175
The paper investigates a possibility of combining the population learning algorithm and the A-Team concept with a view to increase quality of results and efficiency of computations. To implement the idea a middleware environment called JABAT is used. The proposed approach is validated experimentally using benchmark datasets containing instances of the two well-known combinatorial optimization problems: flow shop and job shop scheduling.
The paper focuses on the radial basis function neural design problem. Performance of the RBF neural network strongly depends on the network structure and parameters. Making choices with respect to the structure and va...
详细信息
ISBN:
(数字)9783319198576
ISBN:
(纸本)9783319198576;9783319198569
The paper focuses on the radial basis function neural design problem. Performance of the RBF neural network strongly depends on the network structure and parameters. Making choices with respect to the structure and value of the RBF network parameters involves both stages of its design: initialization and training. The basic question considered in this paper is how these two stages should be carried-out. Should they be carried-out sequentially, in parallel, or perhaps based on other predefined schema or strategy? In the paper an agent-based population learning algorithm is used as a tool for designing of the RBF network. Computational experiment has been planned and executed with a view to investigate effectiveness of different approaches. Experiment results have been analyzed to draw some general conclusions with respect to strategies for the agent-based RBF network design.
The main contribution of the paper is proposing and evaluating, through the computational experiment, an agent-based population learning algorithm generating a representative training dataset of the required size. The...
详细信息
ISBN:
(纸本)9783642219993;9783642220005
The main contribution of the paper is proposing and evaluating, through the computational experiment, an agent-based population learning algorithm generating a representative training dataset of the required size. The proposed approach is based on the assumption that prototypes are selected from clusters. Thus, the number of clusters produced has a direct influence on the size of the reduced dataset. Agents within an A-Team execute various local search procedures and cooperate to find-out a solution to the instance reduction problem aiming at obtaining a compact representation of the dataset. Computational experiment has confirmed that the proposed algorithm is competitive to other approaches.
The problem addressed in this paper concerns data reduction. In the paper the agent-based data reduction algorithm is extended by adding mechanism of integration of the multiple learning models into a single multiple ...
详细信息
ISBN:
(纸本)9783642404955;9783642404948
The problem addressed in this paper concerns data reduction. In the paper the agent-based data reduction algorithm is extended by adding mechanism of integration of the multiple learning models into a single multiple classification system called ensemble model. The paper includes the overview of the proposed approach and discusses the computational experiment results.
暂无评论