Island models meet rumor spreading
6
Citation
23
Reference
10
Related Paper
Citation Trend
Abstract:
Island models in evolutionary computation solve problems by a careful interplay of independently running evolutionary algorithms on the island and an exchange of good solutions between the islands. In this work, we conduct rigorous run time analyses for such island models trying to simultaneously obtain good run times and low communication effort.Keywords:
Rumor
Divide-and-Evolve (DaE) is an original "memeticization" of Evolutionary Computation and Artificial Intelligence Planning. However, like any Evolutionary Algorithm, DaE has several parameters that need to be tuned, and the already excellent experimental results demonstrated by DaE on benchmarks from the International Planning Competition, at the level of those of standard AI planners, have been obtained with parameters that had been tuned once and forall using the Racing method. This paper demonstrates that more specific parameter tuning (e.g. at the domain level or even at the instance level) can further improve DaE results, and discusses the trade-off between the gain in quality of the resulting plans and the overhead in terms of computational cost.
Generality
Cite
Citations (11)
Evolutionary computation (EC) has been a fascinating branch of computation inspiredby a natural phenomenal of evolution. EC enables computer scientists to design eective algorithmsdealing dicult problems. This paper focuses on a special class problem called multi-objective optimizationproblems and evolutionary algorithms designed for it. We will overview the development ofmulti-objective evolutionary algorithms (MOEAs) over the years and problem diculties and thenindicate the open problems in this area. Our chief goal is to provide readers reference material in thearea of multi-objective evolutionary algorithms
Human-based evolutionary computation
Cite
Citations (0)
Association rule analysis has been widely employed as a basic technique for data mining. Extensive research has also been conducted to apply evolutionary computing techniques to the field of data mining. This study presents a method to evaluate the settings of evolutionary operations in evolutionary rule discovery method, which is characterized by the execution of overall problem solving through the acquisition and accumulation of small results. Since the purpose of population evolution is different from that of general evolutionary computation methods that aim at discovering elite individuals, we examined the difference in the concept of settings during evolution and the evaluation of evolutionary computation by visualizing the progress and efficiency of problem solving. The rule discovery method (GNMiner) is characterized by reflecting acquired information in evolutionary operations; this study determines the relationship between the settings of evolutionary operations and the progress of each task execution stage and the achievement of the final result. This study obtains knowledge on the means of setting up evolutionary operations for efficient rule-set discovery by introducing an index to visualize the efficiency of outcome accumulation. This implies the possibility of setting up dynamic evolutionary operations in the outcome accumulation-type evolutionary computation in future studies.
Human-based evolutionary computation
Evolutionary music
Cite
Citations (0)
Evolutionary computation comes from the natural evolution of evolutionary thinking and inspiration. Parallel to its potential and self-organizing, adaptive, self-learning smart features for solving multi-objective optimization problems with great potential. Systematically introduces the multi-objective evolutionary algorithms to optimize multi-objective evolutionary algorithm (MOEAs), at the same time discusses the evolutionary algorithms (EAs) in the multi-objective optimization of the application of a number of key issues in the future and the need for further research work.
Human-based evolutionary computation
Cite
Citations (0)
Some interesting features of the new book "Introduction to Evolutionary Algorithms", which is written by Xinjie Yu and Mitsuo Gen and be published by Springer in 2010, will be illustrated, including covering nearly all the hot evolutionary-computation-related topics, referring to the latest published journal papers, introducing the applications of EAs as many as possible, and adopting many pedagogical ways to make EAs easy and interesting. The contents and the consideration of selecting these contents will be discussed. Then the focus will be put on the pedagogical ways for teaching and self-study. The algorithms introduced in the tutorial include constrained optimization evolutionary algorithms (COEA), multiobjective evolutionary algorithms (MOEA), ant colony optimization (ACO), particle swarm optimization (PSO), and artificial immune systems (AIS). Some of the applications of these evolutionary algorithms will be discussed.
Human-based evolutionary computation
Parallel metaheuristic
Cite
Citations (87)
In this work, we study the performance of an evolutionary algorithm for solving a real-world wind turbine design optimization problem which was a part of the Evolutionary Computation Competition 2019 organized by the Japanese Society of Evolutionary Computation. The problem involves 5 objectives, 32 continuous variables and 22 constraints, which are evaluated using WISDEM and OpenMDAO tools. The results obtained by the proposed algorithm are compared with several state-of-the-art algorithms to demonstrate its effectiveness.
Human-based evolutionary computation
Optimization algorithm
Cite
Citations (0)
Bound‐constrained optimization has wide applications in science and engineering. In the last two decades, various evolutionary algorithms (EAs) were developed under the umbrella of evolutionary computation for solving various bound‐constrained benchmark functions and various real‐world problems. In general, the developed evolutionary algorithms (EAs) belong to nature‐inspired algorithms (NIAs) and swarm intelligence (SI) paradigms. Differential evolutionary algorithm is one of the most popular and well‐known EAs and has secured top ranks in most of the EA competitions in the special session of the IEEE Congress on Evolutionary Computation. In this paper, a customized differential evolutionary algorithm is suggested and applied on twenty‐nine large‐scale bound‐constrained benchmark functions. The suggested C‐DE algorithm has obtained promising numerical results in its 51 independent runs of simulations. Most of the 2013 IEEE‐CEC benchmark functions are tackled efficiently in terms of proximity and diversity.
Benchmark (surveying)
Differential Evolution
Cite
Citations (7)
Human-based evolutionary computation
Margin (machine learning)
Cite
Citations (3)
In the Evolutionary Computation field, it is frequent to assume that a computation load necessary for fitness value computation is, at least, similar for all possible cases. The main objective of this paper is to show that the above assumption is frequently false. Therefore, the examples of evolutionary methods that use problem encoding which allows for significant optimization of the fitness computation process are pointed out and analyzed. The definition of Problem Encoding Allowing Cheap Fitness Computation of Mutated Individuals (PEACh) is proposed. Another objective of the paper is to start a discussion concerning the computation load measurement in the evolutionary computation field. As shown, the Fitness Function Evaluation number is not always a fair measure and may be significantly affected by the quality of method implementation.
Fitness approximation
Human-based evolutionary computation
Value (mathematics)
Cite
Citations (11)
Differential evolution (DE) is an evolutionary algorithm (EA) that uses a rather greedy and less stochastic approach to solve optimization problems than other evolutionary methods [1]. Like other EAs, DE is a population-based, stochastic global optimizer, capable of working reliably in nonlinear and multimodal environments. Due to several features such as simplicity, efficiency and global search capabilities, DE rapidly became a successful paradigm of evolutionary computation. However, to achieve adequate performance with DE, the process of tuning the control parameters is essential as its performance is sensitive to the choice of both mutation and crossover settings. This paper proposes a DE algorithm with adaptive tuning of scaling factor (F), crossover rate (CR) and quasi-oppositional probability based on population's variance information - Adaptive Differential Evolution (ADE). Furthermore, ADE adopts a vector called Fm in each dimension of the optimization problem instead of single variable for F as presented in the classical DE approach. The proposed optimization method is validated on the test-bed proposed for the IEEE CEC'13 (IEEE Congress on Evolutionary Computation 2013) contest for real parameter single objective optimization with 28 benchmark functions. Simulation results over the benchmark functions demonstrate the effectiveness and usefulness of the proposed ADE method. This version of paper includes the ADE's performance on the 10, 30 and 50-dimensional benchmark functions.
Differential Evolution
Benchmark (surveying)
Cite
Citations (39)