In this paper we argue that to produce good optimization performances, the exploration of the solution space does not need to be carried out in the unorderly fashion most evolutionaryalgorithms use. Other strategies ...
详细信息
ISBN:
(纸本)9781479974924
In this paper we argue that to produce good optimization performances, the exploration of the solution space does not need to be carried out in the unorderly fashion most evolutionaryalgorithms use. Other strategies that seek to minimize the cost involved in successive evaluation processes should be explored. This does not imply a fundamental change on how evolutionaryalgorithms work, but rather, it brings some structure onto how solution spaces are explored by contemplating decoding cost as one of the elements to be minimized when sampling. The traditional implementations of most evolutionaryalgorithms assume that any point in the solution space can be evaluated any time and at no cost. However, this is not always the case and often each step of the process only part of the solution space is available for evaluation giving rise to a class of problems we have called constrainedsampling optimization problems over which evolutionaryalgorithms are quite inefficient. To address these problems we have proposed a modification of the general strategy of evolutionaryalgorithms to address these constraints efficiently. Here, we study the effects of this approach when applied to problems that are not constrained, thus modifying the way the solution space is explored. This study is carried out to determine how these modification impact the performance of a set of popular evolutionaryalgorithms over a representative set of benchmark functions corresponding to fitness landscapes with a variety of characteristics. We show that by restricting the sampling capabilities of most algorithms, the cost of the optimization procedure is reduced for most types of fitness landscapes without affecting their results.
In this paper we argue that to produce good optimization performances, the exploration of the solution space does not need to be carried out in the unorderly fashion most evolutionaryalgorithms use. Other strategies ...
详细信息
ISBN:
(纸本)9781479974931
In this paper we argue that to produce good optimization performances, the exploration of the solution space does not need to be carried out in the unorderly fashion most evolutionaryalgorithms use. Other strategies that seek to minimize the cost involved in successive evaluation processes should be explored. This does not imply a fundamental change on how evolutionaryalgorithms work, but rather, it brings some structure onto how solution spaces are explored by contemplating decoding cost as one of the elements to be minimized when sampling. The traditional implementations of most evolutionaryalgorithms assume that any point in the solution space can be evaluated any time and at no cost. However, this is not always the case and often each step of the process only part of the solution space is available for evaluation giving rise to a class of problems we have called constrainedsampling optimization problems over which evolutionaryalgorithms are quite inefficient. To address these problems we have proposed a modification of the general strategy of evolutionaryalgorithms to address these constraints efficiently. Here, we study the effects of this approach when applied to problems that are not constrained, thus modifying the way the solution space is explored. This study is carried out to determine how these modification impact the performance of a set of popular evolutionaryalgorithms over a representative set of benchmark functions corresponding to fitness landscapes with a variety of characteristics. We show that by restricting the sampling capabilities of most algorithms, the cost of the optimization procedure is reduced for most types of fitness landscapes without affecting their results.
In this work we address the solution of a particular category of problems, denoted as constrainedsampling problems, using evolution. These problems have not usually been addressed using EAs. They are characterized by...
详细信息
ISBN:
(纸本)9783642408465
In this work we address the solution of a particular category of problems, denoted as constrainedsampling problems, using evolution. These problems have not usually been addressed using EAs. They are characterized by the fact that the fitness landscape evaluation is not always straightforward due to the computational cost or to physical constraints of the specific application. The decoding phase of these problems usually implies some type of physical migration from the constructs generated to obtain the fitness of the parents towards those required to obtain the fitness of the offspring. As a consequence, it is not instantaneous and requires a series of steps. Most traditional evolutionaryalgorithms ignore the information on the fitness landscape that can be obtained from these intermediate steps. We propose a series of modifications that can be applied to most EAs that allow improving their efficiency when applied to this type of problems. This approach has been tested using some common real-coded benchmark functions and its performance compared to that of a standard EA, specifically a Differential Evolution algorithm.
暂无评论