In our previous work, a multi-objective evolutionary algorithm (MEA_CDPs) was proposed for detecting separated and overlapping communities simultaneously. However, MEA_CDPs has a couple of defects, like individuals ca...
详细信息
In our previous work, a multi-objective evolutionary algorithm (MEA_CDPs) was proposed for detecting separated and overlapping communities simultaneously. However, MEA_CDPs has a couple of defects, like individuals cannot be transformed to community structure by the decoder when the quality of community structure is lower certain thresholds, many vertices with weak overlapping nature are identified as overlapping nodes, and the objective functions can not control the ratio of separated nodes to overlapping nodes. Therefore, in this paper, to overcome these defects, we improve MEA_CDPs by designing more efficient objective functions. We also extend MEA_CDPs' capability in detecting hierarchical community structures. The improved algorithm is named as iMEA_CDPs. In the experiments, a set of computer-generated networks are first used to test the effect of parameters in iMEA_CDPs, and then four real-world networks are used to validate the performance of iMEA_CDPs. The experimental results show that iMEA_CDPs outperforms MEA_CDPs. Moreover, compared with MEA_CDPs, iMEA_CDPs can detect various kinds of overlapping and hierarchical community structures.
A regression tree is a type of decision tree that can be applied to solve regression problems. One of its characteristics is that it may have at least four different node representations;internal nodes can be associat...
详细信息
A regression tree is a type of decision tree that can be applied to solve regression problems. One of its characteristics is that it may have at least four different node representations;internal nodes can be associated with univariate or oblique tests, whereas the leaves can be linked with simple constant predictions or multivariate regression models. The objective of this paper is to demonstrate the impact of particular representations on the induced decision trees. As it is difficult if not impossible to choose the best representation for a particular problem in advance, the issue is investigated using a new evolutionary algorithm for the decision tree induction with a structure that can self-adapt to the currently analyzed data. The proposed solution allows different leaves and internal nodes representation within a single tree. Experiments performed using artificial and real-life datasets show the importance of tree representation in terms of error minimization and tree size. In addition, the presented solution managed to outperform popular tree inducers with defined homogeneous representations. (C) 2016 Elsevier B.V. All rights reserved.
This paper proposes a multi-objective evolutionary algorithm method for Distribution feeder reconfiguration (DFR) with distributed generators (DG) in a practical system. Considering the low inertia constant of DG unit...
详细信息
This paper proposes a multi-objective evolutionary algorithm method for Distribution feeder reconfiguration (DFR) with distributed generators (DG) in a practical system. Considering the low inertia constant of DG units in order to take the transient stability of DGs into account is one of the major issues in power systems. Especially when the penetration of DGs is low, the impacts of them on the distribution system transient stability may be neglected. However, when the penetration of DG increases, the transient stability of them must be taken into account (more DGs, more transient issues). To this end, the DFR problem has been solve by an enhanced Gravitational Search Algorithm (EGSA) to improve the transient stability index and decrease losses and operation cost in a distribution test system with multiple micro-turbines. The effectiveness of the proposed approach is studied based on a typical 33-bus test system. For getting close to the practical condition and considering the detailed dynamic models of the generators and other electric devices in power system, simulation and programming of this approach are done by the DIgSILENT (R) Power Factory software. (C) 2015 Elsevier Ltd. All rights reserved.
This paper describes a new system for automated text clustering, it was specifically designed to work with scientific articles written in Brazilian Portuguese. The system has two modules, the first one extracts the ma...
详细信息
This paper describes a new system for automated text clustering, it was specifically designed to work with scientific articles written in Brazilian Portuguese. The system has two modules, the first one extracts the main terms and generates an index for each text, the second module uses this index to group the texts within their original topics. The first module innovates by selecting compound terms instead of single terms to produce the indexes;the second module applies a new evolutionary clustering algorithm, having a new method of work, over the indexes. The system was tested with four different corpora, the metrics reveal a reasonable clustering result when combining compound terms and the evolutionary method proposed, although the number of clusters generated is distant from the original number of topics. The time consumed by the new clustering algorithm is high when running it over a conventional personal computer.
In this paper a new approach to design sound phase diffusers is presented. The acoustic properties of such diffusers are usually increased by using single objective optimization methods. Here we propose the use of a m...
详细信息
In this paper a new approach to design sound phase diffusers is presented. The acoustic properties of such diffusers are usually increased by using single objective optimization methods. Here we propose the use of a multiobjective (MO) approach to design them in order to take into account several conflicting characteristic simultaneously. Three different MO problems are posed to consider various scenarios where fundamentally the objective is to maximize the normalized diffusion coefficient (following the corresponding Audio Engineering Society standard) for the so-called medium frequencies. This single objective could be divided into other several objectives to adjust performances to designer preferences. A multi-objective evolutionary algorithm (called ev-MOGA) is used to characterize the Pareto front in a smart way. ev-MOGA is modified, by using integer codification and tuning some of its genetic operators, to adapt it to the new requirements. Special interest is posed in selecting the diffusers codification properly to eliminate duplicities that would produce a multimodal problem. Precision in the manufacturing process is taking into account in the diffuser codification causing, that the number of different diffusers are quantified. Robust considerations related with the precision manufacturing process are considered in the decision making process. Finally, an optimal diffuser is selected considering designer preferences.
The iterated prisoner's dilemma is a famous model of cooperation and conflict in game theory. Its origin can be traced back to the Cold War, and countless strategies for playing it have been proposed so far, eithe...
详细信息
The iterated prisoner's dilemma is a famous model of cooperation and conflict in game theory. Its origin can be traced back to the Cold War, and countless strategies for playing it have been proposed so far, either designed by hand or automatically generated by computers. In the 2000s, scholars started focusing on adaptive players, that is, able to classify their opponent's behavior and adopt an effective counter-strategy. The player presented in this paper, pushes such idea even further: it builds a model of the current adversary from scratch, without relying on any pre-defined archetypes, and tweaks it as the game develops using an evolutionary algorithm;at the same time, it exploits the model to lead the game into the most favorable continuation. Models are compact nondeterministic finite state machines;they are extremely efficient in predicting opponents' replies, without being completely correct by necessity. Experimental results show that such a player is able to win several one-to-one games against strong opponents taken from the literature, and that it consistently prevails in round-robin tournaments of different sizes.
This paper studies a family of redundant binary representations NNg(l, k), which are based on the mathematical formulation of error control codes, in particular, on linear block codes, which are used to add redundancy...
详细信息
This paper studies a family of redundant binary representations NNg(l, k), which are based on the mathematical formulation of error control codes, in particular, on linear block codes, which are used to add redundancy and neutrality to the representations. The analysis of the properties of uniformity, connectivity, synonymity, locality and topology of the NNg(l, k) representations is presented, as well as the way an (1+1)-ES can be modeled using Markov chains and applied to NK fitness landscapes with adjacent neighborhood. The results show that it is possible to design synonymously redundant representations that allow an increase of the connectivity between phenotypes. For easy problems, synonymously NNg(l, k) representations, with high locality, and where it is not necessary to present high values of connectivity are the most suitable for an efficient evolutionary search. On the contrary, for difficult problems, NNg(l, k) representations with low locality, which present connectivity between intermediate to high and with intermediate values of synonymity are the best ones. These results allow to conclude that NNg(l, k) representations with better performance in NK fitness landscapes with adjacent neighborhood do not exhibit extreme values of any of the properties commonly considered in the literature of evolutionary computation. This conclusion is contrary to what one would expect when taking into account the literature recommendations. This may help understand the current difficulty to formulate redundant representations, which are proven to be successful in evolutionary computation. (C) 2016 Elsevier B.V. All rights reserved.
One of the most accurate types of prototype selection algorithms, preprocessing techniques that select a subset of instances from the data before applying nearest neighbor classification to it, are evolutionary approa...
详细信息
One of the most accurate types of prototype selection algorithms, preprocessing techniques that select a subset of instances from the data before applying nearest neighbor classification to it, are evolutionary approaches. These algorithms result in very high accuracy and reduction rates, but unfortunately come at a substantial computational cost. In this paper, we introduce a framework that allows to efficiently use the intermediary results of the prototype selection algorithms to further increase their accuracy performance. Instead of only using the fittest prototype subset generated by the evolutionary algorithm, we use multiple prototype subsets in an ensemble setting. Secondly, in order to classify a test instance, we only use prototype subsets that accurately classify training instances in the neighborhood of that test instance. In an experimental evaluation, we apply our new framework to four state-of-the-art prototype selection algorithms and show that, by using our framework, more accurate results are obtained after less evaluations of the prototype selection method. We also present a case study with a prototype generation algorithm, showing that our framework is easily extended to other preprocessing paradigms as well. (C) 2016 Elsevier B.V. All rights reserved.
We report on the application of an evolutionary algorithm (EA) to enhance performance of an ultra-cold quantum gas experiment. The production of a 87 rubidium Bose-Einstein condensate (BEC) can be divided into fundame...
详细信息
We report on the application of an evolutionary algorithm (EA) to enhance performance of an ultra-cold quantum gas experiment. The production of a 87 rubidium Bose-Einstein condensate (BEC) can be divided into fundamental cooling steps, specifically magneto-optical trapping of cold atoms, loading of atoms to a far-detuned crossed dipole trap, and finally the process of evaporative cooling. The EA is applied separately for each of these steps with a particular definition for the feedback, the so-called fitness. We discuss the principles of an EA and implement an enhancement called differential evolution. Analyzing the reasons for the EA to improve, e.g., the atomic loading rates and increase the BEC phase-space density, yields an optimal parameter set for the BEC production and enables us to reduce the BEC production time significantly. Furthermore, we focus on how additional information about the experiment and optimization possibilities can be extracted and how the correlations revealed allow for further improvement. Our results illustrate that EAs are powerful optimization tools for complex experiments and exemplify that the application yields useful information on the dependence of these experiments on the optimized parameters.
Optimization methods are widely used in computational models to improve the outcomes or tuning the model parameters, which often benefit human real life. This thesis introduces a new sampling technique called Opposite...
详细信息
Optimization methods are widely used in computational models to improve the outcomes or tuning the model parameters, which often benefit human real life. This thesis introduces a new sampling technique called Opposite-Center Learning (OCL) intended for convergence speed-up of meta-heuristic optimization algorithms. The simple version of OCL, 1-1 OCL, comprises an extension of Opposition-Based Learning (OBL), a sim- ple scheme that manages to boost numerous optimization methods by considering the opposite points of candidate solutions. In contrast to OBL, 1-1 OCL has a theoretical foundation – the opposite-center point is defined as the optimal choice in pair-wise sam- pling of the search space given a random starting point. A concise analytical background is provided. Based on the research of 1-1 OCL, m-n OCL is developed so that OCL can generate n points from m known points and grant their optimality in the sense of all m and n points generated by m-n OCL scheme having shorter expected distances to an ar- bitrary distributed global optimum. Computationally both the opposite-center point in 1- 1 OCL and opposite-center points in m-n OCL are approximated by a lightweight Monte Carlo scheme for arbitrary dimension. Empirical results up to dimension 20 confirm that 1-1 OCL outperforms OBL and random sampling: the points generated by OCL have shorter expected distances to a uniformly distributed global optimum, where m-n OCL does even better. To further test its practical performance, both 1-1 OCL and m-n OCL are applied to differential evolution (DE). This novel scheme for continuous optimization named Opposite-Center DE (OCDE) and m-n Opposite-Center DE (MNOCDE), they employ OCL for population initialization and generation jumping. Numerical experi- ments on a set of benchmark functions for dimensions 10, 30, 50 and 100 reveal that OCDE and MNOCDE on average improves the convergence rates compared to the orig- inal DE and the Opposition-based DE (ODE), respectively, w
暂无评论