Multiobjective evolutionary algorithms (MOEAs) have witnessed prosperity in solving many-objective optimization problems (MaOPs) over the past three decades. Unfortunately, no one single MOEA equipped with given param...
详细信息
Multiobjective evolutionary algorithms (MOEAs) have witnessed prosperity in solving many-objective optimization problems (MaOPs) over the past three decades. Unfortunately, no one single MOEA equipped with given parameter settings, mating-variation operator, and environmental selection mechanism is suitable for obtaining a set of solutions with excellent convergence and diversity for various types of MaOPs. The reality is that different MOEAs show great differences in handling certain types of MaOPs. Aiming at these characteristics, this paper proposes a flexible ensemble framework, namely, ASES, which is highly scalable for embedding any number of MOEAs to promote their advantages. To alleviate the undesirable phenomenon that some promising solutions are discarded during the evolution process, a big archive that number of contained solutions be far larger than population size is integrated into this ensemble framework to record large-scale nondominated solutions, and also an efficient maintenance strategy is developed to update the archive. Furthermore, the knowledge coming from updating archive is exploited to guide the evolutionary process for different MOEAs, allocating limited computational resources for efficient algorithms. A large number of numerical experimental studies demonstrated superior performance of the proposed ASES. Among 52 test instances, the ASES performs better than all the six baseline algorithms on at least half of the test instances with respect to both metrics hypervolume and inverted generational distance.
The advent of numerical computational approaches permits evolutionary algorithms (EAs) to solve complex, real-world engineering problems. The additional modification or hybridization of such EAs in academic research a...
详细信息
The advent of numerical computational approaches permits evolutionary algorithms (EAs) to solve complex, real-world engineering problems. The additional modification or hybridization of such EAs in academic research and application demonstrates improved performance for domain-specific challenges. However, developing a new algorithm or comparison and selection of existing EAs for challenges in the field of optimization is relatively unexplored. The performance of different well-established algorithms is, therefore, investigated in this work. The selection of algorithms using nonparametric tests encompasses different categories to include- Genetic Algorithm, Particle Swarm Optimization, Harmony Search Algorithm, Cuckoo Search Algorithm, Bat Algorithm, Firefly algorithm, Differential Evolution, and Artificial Bee Colony. These algorithms are applied to solve test functions, including unconstrained, constrained, industry specific problems, CEC 2011 real world optimization problems and selected CEC 2013 benchmark test functions. The three distinct performance metrics, namely, efficiency, reliability, and quality of solution derived using the quantitative attributes are provided to evaluate the performance of the employed EAs. The categorical assignment of performance attributes helps to compare different algorithms for a specific optimization problem while the performance metrics are useful to provide the common platform for new or hybrid EA development.
For general video game playing agents, the biggest challenge is adapting to the wide variety of situations they encounter and responding appropriately. Some success was recently achieved by modifying search-control pa...
详细信息
ISBN:
(纸本)9781728145334
For general video game playing agents, the biggest challenge is adapting to the wide variety of situations they encounter and responding appropriately. Some success was recently achieved by modifying search-control parameters in agents on-line, during one play-through of a game. We propose adapting such methods for Rolling Horizon evolutionary algorithms, which have shown high performance in many different environments, and test the effect of on-line adaptation on the agent's win rate. On-line tuned agents are able to achieve results comparable to the state of the art, including first win rates in hard problems, while employing a more general and highly adaptive approach. We additionally include further insight into the algorithm itself, given by statistics gathered during the tuning process and highlight key parameter choices.
The widespread use of computing devices and the heavy dependence on the internet has evolved the cyberspace to a cyber world - something comparable to an artificial world. This paper focuses on one of the major proble...
详细信息
ISBN:
(纸本)9781728146850
The widespread use of computing devices and the heavy dependence on the internet has evolved the cyberspace to a cyber world - something comparable to an artificial world. This paper focuses on one of the major problems of the cyber world - cyber security or more specifically computer malware. We show that computer malware is a perfect example of an artificial ecosystem with a co-evolutionary predator-prey framework. We attempt to merge the two domains of biologically inspired computing and computer malware. Under the aegis of proactive defense, this paper discusses the possibilities, challenges and opportunities in fusing evolutionary computing techniques with malware creation.
We address the task of repairing infeasibility in the context of infeasible job shop scheduling problems with a hard constraint on the maximum makespan allowed. For this purpose, we adopt a job-based view of repairs, ...
详细信息
We address the task of repairing infeasibility in the context of infeasible job shop scheduling problems with a hard constraint on the maximum makespan allowed. For this purpose, we adopt a job-based view of repairs, that allows for dropping some of the jobs and so gives rise to the problem of computing the largest subset of jobs that can be scheduled under the makespan constraint. Recent work proposed a genetic algorithm for solving this problem, which integrates an efficient solution builder for defining the search space. In this paper, we build on this earlier work and make several contributions. We provide a formal analysis of both the search space and the solution builder. Then, we propose two important enhancements to the genetic algorithm: first, we develop a new solution builder aimed at reducing the number of feasibility tests, making the search process more efficient. In addition, we propose a more effective procedure for testing the feasibility of different subsets of jobs under the given makespan constraint based on the use of a light-weight genetic algorithm. Experimental results show that the proposed methods are effective at solving the problem, and that the enhancements bring significant improvements.
Excel solver is a powerful tool for optimization of linear and nonlinear problems. With this unique tool, the user can achieve an optimal value for the desired objective function in Excel cell. This solver acts on a g...
详细信息
Excel solver is a powerful tool for optimization of linear and nonlinear problems. With this unique tool, the user can achieve an optimal value for the desired objective function in Excel cell. This solver acts on a group of cells that are directly or indirectly associated with the function;thus, the user-defined values will be optimized. In the present work, 13 existing species in an electrolyte solution have been considered to predict the activity coefficient of inorganic ions in the electrolyte solution, which includes H2O, CO2(aq), H+, Na+, Ba2+, Ca2+, Sr2+, Mg2+, OH-, Cl-, SO4, CO3, HCO3. In this study, to predict the activity coefficient of species in the system, Extended UNIQUAC activity coefficient model was considered and its parameters optimized using Excel solver tool based on GRG algorithm. Total error for optimization of adjustable parameters of Extended UNIQUAC model for 13 desired ions at three temperatures 298.15 K, 323.15 K and 373.15 K with the Excel solver tool was 0.0087. The results of GRG algorithm were favorable than those of ICA, PSO and ABC algorithms. The results of this optimization are intended to predict mineral deposition. The number of adjustable variables (model parameters for optimization) is over 200, and the number of target functions is 39.
It has been widely recognized that evolutionary computation is one of the most effective techniques for solving complex optimization problems. As a group of meta-heuristics inspired by nature, the superiority of evolu...
详细信息
ISBN:
(纸本)9781728169293
It has been widely recognized that evolutionary computation is one of the most effective techniques for solving complex optimization problems. As a group of meta-heuristics inspired by nature, the superiority of evolutionary algorithms is mainly attributed to the evolution of multiple candidate solutions, which can strike a balance between exploration and exploitation. However, the effectiveness of evolutionary algorithms is generally at the expense of efficiency, which reduces the prevalence of evolutionary algorithms in solving real-world optimization problems. In 2017, we proposed the evolutionary multi-objective optimization platform PlatEMO to facilitate the use of multi-objective evolutionary algorithms (MOEAs), where some delicate techniques were developed to improve the computational efficiency of MOEAs. These techniques have not been introduced before, since users need not care about them when using existing MOEAs or developing new MOEAs. To deepen the understanding of the core mechanisms of PlatEMO, this paper gives a comprehensive introduction to these techniques, including new non-dominated sorting approaches, matrix calculation, and parallel computing. Several comparative experiments are conducted for a quantitative understanding of the efficiency improvement brought by these techniques.
This paper proposes a new parent selection strategy for reducing the effect of evaluation time bias in asynchronous parallel multi-objective evolutionary algorithms. Asynchronous parallel evolutionary algorithms have ...
详细信息
ISBN:
(纸本)9781728125473
This paper proposes a new parent selection strategy for reducing the effect of evaluation time bias in asynchronous parallel multi-objective evolutionary algorithms. Asynchronous parallel evolutionary algorithms have a problem to biased toward the search region with short evaluation time. The proposed method selects parent solutions that take into account the search progress of each solution. This paper conducts an experiment to investigate the effectiveness of the proposed method. The experiment uses NSGA-HI. a multi-objective evolutionary algorithm, and compares the synchronous NSGA-III, the asynchronous NSGA-III, and the asynchronous NSGA-III with the proposed selection method. The experimental result reveals that the proposed method can reduce the effect of the evaluation time bias while reducing the computing time of the parallel NSGA-III.
Many important problems can be regarded as maximizing submodular functions under some constraints. A simple multi-objective evolutionary algorithm called GSEMO has been shown to achieve good approximation for submodul...
详细信息
ISBN:
(纸本)9783030581145;9783030581152
Many important problems can be regarded as maximizing submodular functions under some constraints. A simple multi-objective evolutionary algorithm called GSEMO has been shown to achieve good approximation for submodular functions efficiently. While there have been many studies on the subject, most of existing run-time analyses for GSEMO assume a single cardinality constraint. In this work, we extend the theoretical results to partition matroid constraints which generalize cardinality constraints, and show that GSEMO can generally guarantee good approximation performance within polynomial expected run time. Furthermore, we conducted experimental comparison against a baseline GREEDY algorithm in maximizing undirected graph cuts on random graphs, under various partition matroid constraints. The results show GSEMO tends to outperform GREEDY in quadratic run time.
We propose a new class of multi-objective benchmark problems on which we analyse the performance of four well established multi-objective evolutionary algorithms (MOEAs) - each implementing a different search paradigm...
详细信息
ISBN:
(纸本)9783030581114;9783030581121
We propose a new class of multi-objective benchmark problems on which we analyse the performance of four well established multi-objective evolutionary algorithms (MOEAs) - each implementing a different search paradigm - by comparing run-time convergence behaviour over a set of 1200 problem instances. The new benchmarks are created by fusing previously proposed single-objective interpolated continuous optimisation problems (ICOPs) via a common set of Pareto non-dominated seeds. They thus inherit the ICOP property of having tunable fitness landscape features. The benchmarks are of intrinsic interest as they derive from interpolation methods and so can approximate general problem instances. This property is revealed to be of particular importance as our extensive set of numerical experiments indicates that choices pertaining to (i) the weighting of the inverse distance interpolation function and (ii) the problem dimension can be used to construct problems that are challenging to all tested multi-objective search paradigms. This in turn means that the new multi-objective ICOPs problems (MO-ICOPs) can be used to construct well-balanced benchmark sets that discriminate well between the run-time convergence behaviour of different solvers.
暂无评论