A Posteriori Approach for Community Detection

Conventional community detection approaches in complex network are based on the optimization of a priori decision, i.e., a single quality function designed beforehand. This paper proposes a posteriori decision approach for community detection. The approach includes two phases: in the search phase, a special multi-objective evolutionary algorithm is designed to search for a set of tradeoff partitions that reveal the community structure at different scales in one run; in the decision phase, three model selection criteria and the Possibility Matrix method are proposed to aid decision makers to select the preferable solutions through differentiating the set of optimal solutions according to their qualities. The experiments in five synthetic and real social networks illustrate that, in one run, our method is able to obtain many candidate solutions, which effectively avoids the resolution limit existing in priori decision approaches. In addition, our method can discover more authentic and comprehensive community structures than those priori decision approaches.     

A hybrid multi-population framework for dynamic environments combining online and offline learning

Population based incremental learning algorithms and selection hyper-heuristics are highly adaptive methods which can handle different types of dynamism that may occur while a given problem is being solved. In this study, we present an approach based on a multi-population framework hybridizing these methods to solve dynamic environment problems. A key feature of the hybrid approach is the utilization of offline and online learning methods at successive stages. The performance of our approach along with the influence of different heuristic selection methods used within the selection hyper-heuristic is investigated over a range of dynamic environments produced by a well known benchmark generator as well as a real world problem, referred to as the Unit Commitment Problem. The empirical results show that the proposed approach using a particular hyper-heuristic outperforms some of the best known approaches in literature on the dynamic environment problems dealt with.     

A goal programming-TOPSIS approach to multiple response optimization using the concepts of non-dominated solutions and prediction intervals

Multiple response problems include three stages: data gathering, modeling and optimization. Most approaches to multiple response optimization ignore the effects of the modeling stage; the model is taken as given and subjected to multi-objective optimization. Moreover, these approaches use subjective methods for the trade off between responses to obtain one or more solutions. In contradistinction, in this paper we use the Prediction Intervals (PIs) from the model building stage to trade off between responses in an objective manner. Our new method combines concepts from the goal programming approach with normalization based on negative and positive ideal solutions as well as the use of prediction intervals for obtaining a set of non-dominated and efficient solutions. Then, the non-dominated solutions (alternatives) are ranked by the TOPSIS (Technique for Order Preference by Similarity to the Ideal Solution) approach. Since some suggested settings of the input variables may not be possible in practice or may lead to unstable operating conditions, this ranking can be extremely helpful to Decision Makers (DMs). The consideration of statistical results together with the selection of the preferred solution among the efficient solutions by Multiple Attribute Decision Making (MADM) distinguishes our approach from others in the literature. We also show, through a numerical example, how the solutions of other methods can be obtained by modifying the relevant approach according to the DM’s requirements.     

Incorporation of implicit decision-maker preferences in multi-objective evolutionary optimization using a multi-criteria classification method

Nowadays, most Multi-Objective Evolutionary Algorithms (MOEA) concentrate mainly on searching for an approximation of the Pareto frontier to solve a multi-objective optimization problem. However, finding this set does not completely solve the problem. The decision-maker (DM) still has to choose the best compromise solution from that set. But as the number of criteria increases, several important difficulties arise in performing this task. Identifying the Region of Interest (ROI), according to the DM’s preferences, is a promising alternative that would facilitate the selection process. This paper approaches the incorporation of preferences into a MOEA in order to characterize the ROI by a multi-criteria classification method. This approach is called Hybrid Multi-Criteria Sorting Genetic Algorithm and is composed of two phases. First, a metaheuristic is used to generate a small set of solutions that are classified in ordered categories by the DM. Thus, the DM’s preferences will be reflected indirectly in this set. In the second phase, a multi-criteria sorting method is combined with an evolutionary algorithm. The first one is used to classify new solutions. Those classified as ‘satisfactory’ are used for creating a selective pressure towards the ROI. The effectiveness of our method was proved in nine instances of a public project portfolio problem. The obtained results indicate that our approach achieves a good characterization of the ROI, and outperforms the standard NSGA-II in simple and complex problems. Also, these results confirm that our approach is able to deal with many-objective problems.     

A blackboard control architecture for model selection and sequencing

Model management systems are computerised systems that facilitate the management of large numbers of decision models used in organizations. Model selection and sequencing in a model management system is the problem of processing a given model base in order to arrive at a sequence of models that can be executed to produce a set of required outputs (goal). Prior solution approaches do not attempt to solve this problem such that the goal is achieved while best meeting the objectives of the user. Instead, research to date has typically provided the first sequence of models which satisfy the goal, without attempting to optimise the objectives of the user. This restricts the applicability of many existing approaches to problems with unique solutions or to situations where users exhibit no preference among the candidate model sequences (i.e. solutions). In many real-world problems, however, multiple solutions may exist and users may prefer a certain solution over the others, based on a variety of criteria such as solution cost, accuracy and so on. In this paper, we present an architecture based on the concept of blackboard control that solves the model selection and sequencing problem while attempting to optimise the objectives of the user. We also discuss the applicability of the proposed approach for solving other problems encountered in the area of model management.     

A hybrid Hopfield network-genetic algorithm approach for the terminal assignment problem

This paper presents a hybrid Hopfield network-genetic algorithm (GA) approach to tackle the terminal assignment (TA) problem. TA involves determining minimum cost links to form a communications network, by connecting a given set of terminals to a given collection of concentrators. Some previous approaches provide very good results if the cost associated with assigning a single terminal to a given concentrator is known. However, there are situations in which the cost of a single assignment is not known in advance, and only the cost associated with feasible solutions can be calculated. In these situations, previous algorithms for TA based on greedy heuristics are no longer valid, or fail to get feasible solutions. Our approach involves a Hopfield neural network (HNN) which manages the problem's constraints, whereas a GA searches for high quality solutions with the minimum possible cost. We show that our algorithm is able to achieve feasible solutions to the TA in instances where the cost of a single assignment in not known in advance, improving the results obtained by previous approaches. We also show the applicability of our approach to other problems related to the TA.     

Sequential approaches for learning datum-wise sparse representations

In supervised classification, data representation is usually considered at the dataset level: one looks for the ??best?? representation of data assuming it to be the same for all the data in the data space. We propose a different approach where the representations used for classification are tailored to each datum in the data space. One immediate goal is to obtain sparse datum-wise representations: our approach learns to build a representation specific to each datum that contains only a small subset of the features, thus allowing classification to be fast and efficient. This representation is obtained by way of a sequential decision process that sequentially chooses which features to acquire before classifying a particular point; this process is learned through algorithms based on Reinforcement Learning. The proposed method performs well on an ensemble of medium-sized sparse classification problems. It offers an alternative to global sparsity approaches, and is a natural framework for sequential classification problems. The method extends easily to a whole family of sparsity-related problem which would otherwise require developing specific solutions. This is the case in particular for cost-sensitive and limited-budget classification, where feature acquisition is costly and is often performed sequentially. Finally, our approach can handle non-differentiable loss functions or combinatorial optimization encountered in more complex feature selection problems.     

An Efficient Algorithm for Solving Fuzzy Linear Programming Problems

In this article, we consider some well-known approaches for solving fuzzy linear programming (FLP) problems. We present some of the difficulties of these approaches. Then, crisp linear programming problems are suggested for solving FLP problems. A new algorithm is also given. The proposed approach has advantages over the other methods. For example, we can achieve higher membership degrees for objective function and constraints. Moreover, we show that the fuzzy optimal solutions obtained by the proposed approach are efficient enough. Also, we see that unlike the previous methods, our method finds efficient solutions by solving only one crisp linear problem instead of solving two or three crisp problems. Finally some numerical examples are presented to show the efficiency of the given approach over the other approaches.     

An effective coding approach for multiobjective integrated resource selection and operation sequences problem

In this paper, we consider an integrated Resource Selection and Operation Sequences (iRS/OS) problem in Intelligent Manufacturing System (IMS). Several kinds of objectives are taken into account, in which the makespan for orders should be minimized; workloads among machine tools should be balanced; the total transition times between machines in a local plant should also be minimized. To solve this multiobjective iRS/OS model, a new two vectors-based coding approach has been proposed to improve the efficiency by designing a chromosome containing two kinds of information, i.e., operation sequences and machine selection. Using such kind of chromosome, we adapt multistage operation-based Genetic Algorithm (moGA) to find the Pareto optimal solutions. Moreover a special technique called left-shift hillclimber has been used as one kind of local search to improve the efficiency of our algorithm. Finally, the experimental results of several iRS/OS problems indicate that our proposed approach can obtain best solutions. Further more comparing with previous approaches, moGA performs better for finding Pareto solutions. Received: May 2005/Accepted: December 2005     

Solution of Jiles-Atherton vector hysteresis parameters estimation by modified Differential Evolution approaches

Differential Evolution (DE) is a simple and efficient stochastic global optimization algorithm of evolutionary computation field, which involves the evolution of a population of solutions using operators such as mutation, crossover, and selection. The basic idea of DE is to adapt the search during the evolutionary process. At the start of the evolution, the perturbations are large since parent populations are far away from each other. As the evolutionary process matures, the population converges to a small region and the perturbations adaptively become small. DE approaches have been successfully applied to solve a wide range of optimization problems. In this paper, the parameters set of the Jiles-Atherton vector hysteresis model is obtained with an approach based on modified Differential Evolution (MDE) approaches using generation-varying control parameters based on generation of random numbers with uniform distribution. Several evaluated MDE approaches perform better than the classical DE methods and a genetic algorithm approach in terms of the quality and stability of the final solutions in optimization of vector Jiles-Atherton vector hysteresis model from a workbench containing a rotational single sheet tester.     

Primal explicit max margin feature selection for nonlinear support vector machines

Embedding feature selection in nonlinear support vector machines (SVMs) leads to a challenging non-convex minimization problem, which can be prone to suboptimal solutions. This paper develops an effective algorithm to directly solve the embedded feature selection primal problem. We use a trust-region method, which is better suited for non-convex optimization compared to line-search methods, and guarantees convergence to a minimizer. We devise an alternating optimization approach to tackle the problem efficiently, breaking it down into a convex subproblem, corresponding to standard SVM optimization, and a non-convex subproblem for feature selection. Importantly, we show that a straightforward alternating optimization approach can be susceptible to saddle point solutions. We propose a novel technique, which shares an explicit margin variable to overcome saddle point convergence and improve solution quality. Experiment results show our method outperforms the state-of-the-art embedded SVM feature selection method, as well as other leading filter and wrapper approaches.     

Efficient and reliable service selection for heterogeneous distributed software systems

The service-oriented paradigm is emerging as a new approach to heterogeneous distributed software systems composed of services accessed locally or remotely by middleware technology. How to select the optimal composited service from a set of functionally equivalent services with different quality of service (QoS) attributes has become an active focus of research in the service community. However, existing middleware solutions or approaches are inefficient as they search all solution spaces. More importantly, they inherently neglect QoS uncertainty owing to the dynamic network environment. In this paper, based on a service composition middleware framework, we propose an efficient and reliable service selection approach that attempts to select the best reliable composited service by filtering low-reliability services through the computation of QoS uncertainty. The approach first employs information theory and probability theory to abandon high-QoS-uncertainty services and downsize the solution space. A reliability fitness function is then designed to select the best reliable service for composited services. We experimented with real-world and synthetic datasets and compared our approach with other approaches. Our results show that our approach is not only fast, but also finds more reliable composited services.     

基于PE静态结构特征的恶意软件检测方法

针对现有检测方法的不足,提出了一种通过挖掘PE文件结构信息来检测恶意软件的方法,并用最新的PE格式恶意软件进行了实验。结果显示,该方法以99.1%的准确率检测已知和未知的恶意软件,评价的重要指标AUC值是0.998,已非常接近最优值1,高于现有的静态检测方法。同时,与其他方法相比,该检测方法的处理时间和系统开销也是较少的,对采用加壳和混淆技术的恶意软件也保持稳定有效,已达到了实时部署使用要求。此外,现有的基于数据挖掘的检测方法在特征选择时存在过度拟合数据的情况,而该方法在这方面具有较强的鲁棒性。     

EFFICIENT ABSTRACTION SELECTION IN REINFORCEMENT LEARNING

This article addresses reinforcement learning problems based on factored Markov decision processes (MDPs) in which the agent must choose among a set of candidate abstractions, each build up from a different combination of state components. We present and evaluate a new approach that can perform effective abstraction selection that is more resource‐efficient and/or more general than existing approaches. The core of the approach is to make selection of an abstraction part of the learning agent's decision‐making process by augmenting the agent's action space with internal actions that select the abstraction it uses. We prove that under certain conditions this approach results in a derived MDP whose solution yields both the optimal abstraction for the original MDP and the optimal policy under that abstraction. We examine our approach in three domains of increasing complexity: contextual bandit problems, episodic MDPs, and general MDPs with context‐specific structure. © 2013 Wiley Periodicals, Inc.     

IFS-CoCo: Instance and feature selection based on cooperative coevolution with nearest neighbor rule

Feature and instance selection are two effective data reduction processes which can be applied to classification tasks obtaining promising results. Although both processes are defined separately, it is possible to apply them simultaneously.This paper proposes an evolutionary model to perform feature and instance selection in nearest neighbor classification. It is based on cooperative coevolution, which has been applied to many computational problems with great success.The proposed approach is compared with a wide range of evolutionary feature and instance selection methods for classification. The results contrasted through non-parametric statistical tests show that our model outperforms previously proposed evolutionary approaches for performing data reduction processes in combination with the nearest neighbor rule.     

An investigation of fuzzy multiple heuristic orderings in the construction of university examination timetables

In this paper, we present an investigation into using fuzzy methodologies to guide the construction of high quality feasible examination timetabling solutions. The provision of automated solutions to the examination timetabling problem is achieved through a combination of construction and improvement. The enhancement of solutions through the use of techniques such as metaheuristics is, in some cases, dependent on the quality of the solution obtained during the construction process. With a few notable exceptions, recent research has concentrated on the improvement of solutions as opposed to focusing on investigating the ‘best’ approaches to the construction phase. Addressing this issue, our approach is based on combining multiple criteria in deciding on how the construction phase should proceed. Fuzzy methods were used to combine three single construction heuristics into three different pair wise combinations of heuristics in order to guide the order in which exams were selected to be inserted into the timetable solution. In order to investigate the approach, we compared the performance of the various heuristic approaches with respect to a number of important criteria (overall cost penalty, number of skipped exams, number of iterations of a rescheduling procedure required and computational time) on 12 well-known benchmark problems. We demonstrate that the fuzzy combination of heuristics allows high quality solutions to be constructed. On one of the 12 problems, we obtained lower penalty than any previously published constructive method and for all 12 we obtained lower penalty than when any of the single heuristics were used alone. Furthermore, we demonstrate that the fuzzy approach used less backtracking when constructing solutions than any of the single heuristics. We conclude that this novel fuzzy approach is a highly effective method for heuristically constructing solutions and, as such, has particular relevance to real-world situations in which the construction of feasible solutions is often a difficult task in its own right.     

An island model genetic algorithm for unequal area facility layout problems

The Unequal Area Facility Layout Problem (UA–FLP) has been addressed by various methods, including mathematical modelling, heuristic and metaheuristic approaches. Nevertheless, each type of approach presents problems such as premature convergence, lack of diversity, or high computational cost. In this paper, for the first time, an Island Model Genetic Algorithm (IMGA) is proposed to solve these subjects in the UA–FLP. The parallel evolution of several populations is used to maintain the population diversity and to obtain a wider sampling of the search space to obtain better quality solutions in fewer generations. Our novel approach was tested with a well-known set of problems taken from the literature and the results were compared with those of previous reports. In most cases, the results obtained by our novel approach improved on the previous results. Additionally, the proposed approach is able to reach good solutions with a wide range of problem sizes and in a reasonable computational time.     

Optimal price zones of electricity markets: a mixed-integer multilevel model and global solution approaches

Mathematical modelling of market design issues in liberalized electricity markets often leads to mixed-integer nonlinear multilevel optimization problems for which no general-purpose solvers exist and which are intractable in general. In this work, we consider the problem of splitting a market area into a given number of price zones such that the resulting market design yields welfare-optimal outcomes. This problem leads to a challenging multilevel model that contains a graph-partitioning problem with multi-commodity flow connectivity constraints and nonlinearities due to proper economic modelling. Furthermore, it has highly symmetric solutions. We develop different problem-tailored solution approaches. In particular, we present an extended Karush-Kuhn-Tucker (KKT) transformation approach as well as a generalized Benders approach that both yield globally optimal solutions. These methods, enhanced with techniques such as symmetry breaking and primal heuristics, are evaluated in detail on academic as well as on realistic instances. It turns out that our approaches lead to effective solution methods for the difficult optimization tasks presented here, where the problem-specific generalized Benders approach performs considerably better than the methods based on KKT transformation.     

An efficient heuristic for series–parallel redundant reliability problems

This paper presents a novel algorithm for solving a series–parallel redundancy allocation problem with separable constraints. The idea of a heuristic approach design is inspired from the greedy method and the genetic algorithm. The structure of the algorithm includes: (1) randomly generating a specified population size number of minimum workable solutions; (2) assigning components either according to the greedy method or to the random selection method; and (3) improving solutions through an inner-system and inter-system solution revision process. Numerical results for the 33 test problems from previous research are reported and compared. As reported in this paper, the solutions found by our approach are all better than or are in par with the well-known best solutions from the approach taken by previous solutions.