An evolutionary algorithm is used to evolve a digital circuit which computes a simple hash function mapping a 16-bit address space into an 8-bit one. The target technology is FPGA, where the search space of the algori...
详细信息
An evolutionary algorithm is used to evolve a digital circuit which computes a simple hash function mapping a 16-bit address space into an 8-bit one. The target technology is FPGA, where the search space of the algorithm is made of the combinational functions computed by cells and of the interconnections among cells. The evolutionary technique has been applied to five different interconnection topologies, specified by neighbourhood graphs. This circuit is readily applicable to the design of set-associative cache memories. Possible use of the evolutionary approach presented in the paper for on-line tuning of the function during cache operation is also discussed.
Recent research shows that orthogonal array based crossovers outperform standard and existing crossovers in evolutionary algorithms in solving parametrical problems with high dimensions and multi-optima. However, thos...
详细信息
Recent research shows that orthogonal array based crossovers outperform standard and existing crossovers in evolutionary algorithms in solving parametrical problems with high dimensions and multi-optima. However, those crossovers employed so far, ignore the consideration of interactions between genes. In this paper, we propose a method to improve the existing orthogonal array based crossovers by integrating information of interactions between genes. It is empirically shown that the proposed orthogonal array based crossover outperforms significantly both the existing orthogonal array based crossovers and standard crossovers on solving parametrical benchmark functions that interactions exist between variables. To further compare the proposed orthogonal array based crossover with the existing crossovers in evolutionary algorithms, a validation test based on car door design is used in which the effectiveness of the proposed orthogonal array based crossover is studied. (C) 2009 Elsevier Ltd. All rights reserved.
Biological networks are structurally adaptive and take on non-random topological properties that influence system robustness. Studies are only beginning to reveal how these structural features emerge, however the infl...
详细信息
Biological networks are structurally adaptive and take on non-random topological properties that influence system robustness. Studies are only beginning to reveal how these structural features emerge, however the influence of component fitness and community cohesion (modularity) have attracted interest from the scientific community. In this study, we apply these concepts to an evolutionary algorithm and allow its population to self-organize using information that the population receives as it moves over a fitness landscape. More precisely, we employ fitness and clustering based topological operators for guiding network structural dynamics, which in turn are guided by population changes taking place over evolutionary time. To investigate the effect on evolution, experiments are conducted on six engineering design problems and six artificial test functions and compared against cellular genetic algorithms and panmictic evolutionary algorithm designs. Our results suggest that a self-organizing topology evolutionary algorithm can exhibit robust search behavior with strong performance observed over short and long time scales. More generally, the coevolution between a population and its topology may constitute a promising new paradigm for designing adaptive search heuristics.
In this paper we introduce some techniques for the analysis of time complexity of evolutionary algorithms (EAs) based on a finite search space. The Markov property and the decomposition of a matrix are employed for th...
详细信息
In this paper we introduce some techniques for the analysis of time complexity of evolutionary algorithms (EAs) based on a finite search space. The Markov property and the decomposition of a matrix are employed for the exact analytic expressions of the mean first hitting times that EAs reach the optimal solutions (FHT-OS). Dynkin's formula, one-step increment analysis and some other probability methods are adopted for the lower and upper bounds of the mean FHT-OS. The theory of a non-negative matrix, including the Perron-Frobenius theorem, the Jordan standard form, etc., are applied for the convergence of EAs. In addition, some techniques for time complexity and convergence of general adaptive EAs are also involved in this paper. Finally, the theoretical results obtained in this paper are verified effectively by applying them to the analysis of a typical evolutionary algorithm, (1 + 1)-EA.
Over the last decade, a variety of evolutionary algorithms (EAs) have been proposed for solving multiobjective optimization problems. Especially more recent multiobjective evolutionary algorithms (MOEAs) have been sho...
详细信息
Over the last decade, a variety of evolutionary algorithms (EAs) have been proposed for solving multiobjective optimization problems. Especially more recent multiobjective evolutionary algorithms (MOEAs) have been shown to be efficient and superior to earlier approaches. In the development of new MOEAs, the strive is to obtain increasingly better performing MOEAs. An important question however is whether we can expect such improvements to converge onto a specific efficient MOEA that behaves best on a large variety of problems. The best MOEAs to date behave similarly or are individually preferable with respect to different performance indicators. In this paper, we argue that the development of new MOEAs cannot converge onto a single new most efficient MOEA because the performance of MOEAs shows characteristics of multiobjective problems. While we will point out the most important aspects for designing competent MOEAs in this paper, we will also indicate the inherent multiobjective tradeoff in multiobjective optimization between proximity and diversity preservation. We will discuss the impact of this tradeoff on the concepts and design of exploration and exploitation operators. We also present a general framework for competent MOEAs and show how current state-of-the-art MOEAs can be obtained by making choices within this framework. Furthermore, we show an example of how we can separate nondomination selection pressure from diversity preservation selection pressure and discuss the impact of changing the ratio between these components.
There is an increasing interest in the application of evolutionary algorithms (EAs) to induce classification rules. This hybrid approach can benefit areas where classical methods for rule induction have not been very ...
详细信息
There is an increasing interest in the application of evolutionary algorithms (EAs) to induce classification rules. This hybrid approach can benefit areas where classical methods for rule induction have not been very successful. One example is the induction of classification rules in imbalanced domains. Imbalanced data occur when one or more classes heavily outnumber other classes. Frequently, classical machine learning (ML) classifiers are not able to learn in the presence of imbalanced data sets, inducing classification models that always predict the most numerous classes. In this work, we propose a novel hybrid approach to deal with this problem. We create several balanced data sets with all minority class cases and a random sample of majority class cases. These balanced data sets are fed to classical ML systems that produce rule sets. The rule sets are combined creating a pool of rules and an EA is used to build a classifier from this pool of rules. This hybrid approach has some advantages over undersampling, since it reduces the amount of discarded information, and some advantages over oversampling, since it avoids overfitting. The proposed approach was experimentally analysed and the experimental results show an improvement in the classification performance measured as the area under the receiver operating characteristics (ROC) curve.
Hundreds of variants of Swarm Intelligence or evolutionary algorithms are proposed each year and numerous competitions and comparisons between algorithms may suggest rapid improvement in the field. However, such compa...
详细信息
Hundreds of variants of Swarm Intelligence or evolutionary algorithms are proposed each year and numerous competitions and comparisons between algorithms may suggest rapid improvement in the field. However, such comparisons are often done between a limited number of methods and are based on averaged ranks of algorithms. This way they measure whether one method is on average ranked better than the others, without giving any information on how much improvement is in fact obtained. In this study we show a general comparison between 69 algorithms, starting from methods proposed in the 1960's up to variants developed in the early 2020's, on single-objective static numerical problems. algorithms are compared on searching for a minimum of 30 different 50-dimensional mathematical functions, and on 22 real-world problems. We focus on the relative improvement achieved by various algorithms over a single-solution based method proposed in 1960 by Howard Rosenbrock. We find that the general improvement of evolutionary algorithms over Rosenbrock's algorithm is relatively limited. It is high for the artificial benchmarks, for which many evolutionary algorithms find solutions 10 times closer to the global optimum in terms of fitness than Rosenbrock's algorithm, but much lower for real-world problems. Improvement is also higher when performance averaged over many runs is compared, but lower when the best results from multiple runs are analyzed. In the last case, only the best evolutionary algorithms are able to find solutions of a "typical " real-world problem that are 2-3 times better in terms of fitness than those found by Rosenbrock's algorithm. The relative improvement of recently proposed algorithms is not much better than the improvement achieved by algorithms proposed over a decade ago.
The flight control system is a key element to achieve required performance in missiles and aircrafts. First purpose of flight control system is to ensuring the stability of the system, then, it attempts to force it to...
详细信息
The flight control system is a key element to achieve required performance in missiles and aircrafts. First purpose of flight control system is to ensuring the stability of the system, then, it attempts to force it to track the guidance commands. This paper provides a straightforward method using evolutionary optimisation algorithms to design an appropriate autopilot for non-minimum phase missiles. In order to bring the results to the actual conditions, the missile non-minimum phase model and actuator dynamics with time delay is considered. Proper indices such as system speed, overshoot, undershoot, steady state error and control signal effort have been incorporated to propose an innovative cost function. Then, several applicable meta-heuristic techniques are employed to optimise this cost function. Genetic algorithm, particle swarm optimisation, artificial bee colony, imperialist competitive algorithm and cuckoo search techniques have been compared in this optimisation problem. Simulation results on two benchmark problem show that this method has acceptable speed and it can be used in gain scheduling control design method for non-minimum phase systems. This method can be a suitable replacement for the time consuming procedure of gain tuning in gain scheduling method. The superior advantage of this method compared to the other methods is automatic tuning of the autopilot gains.
The interest in understanding nanofluid density's impact on heat transfer and fluid flow behaviors has driven the need for accurate density values. Artificial intelligence techniques for predicting nanofluid densi...
详细信息
The interest in understanding nanofluid density's impact on heat transfer and fluid flow behaviors has driven the need for accurate density values. Artificial intelligence techniques for predicting nanofluid density provide a cost-effective and efficient alternative to labor-intensive lab experiments. In the current research, four distinct models based on the Radial Basis Function (RBF) Neural Network were developed and implemented on an extensive and comprehensive databank comprising 4004 experimental data-points gathered from multiple available sources. Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Imperialist Competitive Algorithm (ICA), and Genetic Algorithm (GA) were used separately to optimize the neural network. The provided databank introduces 95 varieties of mono-nanofluids, including 16 types of nano-particles and 11 types of base- fluids. The target/dependent variable in this research is the density of mono-nanofluids (rho nf), whereas the input/ independent variables include the average nano-particle diameter (dnp), nano-particle mass concentration (phi m), temperature (T), pressure (P), nano-particle density (rho np), and base-fluid density (rho bf). Various analyses of the four models confirmed the RBF-ACO model's robustness and superiority. Key statistical indicators for this model indicated an Average Absolute Percent Relative Error (AAPRE) of 0.5391%, a Standard Deviation (SD) of 0.0099, and a Coefficient of Determination (R2) of 0.9818. Sensitivity analysis for the superior model identified key input-output variables with high relevancy factors (r-values). Notably, variables phi m and rho bf had maximum rvalues close to 0.70, indicating their significant role in predicting mono-nanofluids' density. In addition, a leverage statistical approach was utilized to determine possible outliers and the applicability domain of the RBFACO model.
evolutionary algorithms (EAs) excel in optimizing systems with a large number of variables. Previous mathematical and empirical studies have shown that opposition-based algorithms can improve EA performance. We review...
详细信息
evolutionary algorithms (EAs) excel in optimizing systems with a large number of variables. Previous mathematical and empirical studies have shown that opposition-based algorithms can improve EA performance. We review existing opposition-based algorithms and introduce a new one. The proposed algorithm is named fitness-based quasi-reflection and employs the relative fitness of solution candidates to generate new individuals. We provide the probabilistic analysis to prove that among all the opposition-based methods that we investigate, fitness-based quasi-reflection has the highest probability of being closer to the solution of an optimization problem. We support our theoretical findings via Monte Carlo simulations and discuss the use of different reflection weights. We also demonstrate the benefits of fitness-based quasi-reflection on three state-of-the-art EAs that have competed at IEEE CEC competitions. The experimental results illustrate that fitness-based quasi-reflection enhances EA performance, particularly on problems with more challenging solution spaces. We found that competitive DE (CDE) which was ranked tenth in CEC 2013 competition benefited the most from opposition. CDE with fitness-based quasi-reflection improved on 21 out of the 28 problems in the CEC 2013 test suite and achieved 100% success rate on seven more problems than CDE. (C) 2015 Elsevier Ltd. All rights reserved.
暂无评论