Smart use of computational resources based on contribution for cooperative co-evolutionary algorithms

Abstract

Standard Cooperative Co-evolution uses a round-robin method to select subcomponents to undergo optimization. In a non-separable (epistatic) optimization problem, dividing the computational budget equally between all of the subcomponents is not necessarily the best strategy. When dealing with non-separable problems, there is usually an imbalance between the contribution of various subcomponents to the global fitness of the individuals. Using a round-robin fashion treats all of the subcomponents equally and wastes the computational budget. In this paper, we propose a Contribution Based Cooperative Co-evolution (CBCC) that selects the subcomponents based on their contributions to the global fitness. This alleviates the imbalance issue and allows the computational resources to be used more efficiently. Experiments on several benchmark functions with the "imbalance issue" show that this new scheme is promising, especially when it is combined with a grouping algorithm that captures interacting variables in common subcomponents.

Cooperative Co-Evolution With Differential Grouping for Large Scale Optimization

“…It has been shown recently that putting equal emphasis on all the subcomponents in a CC framework is not a very efficient use of the computational budget [15]. Unlike traditional CC, in contribution-based cooperative co-evolution (CBCC) [15], subcomponents are chosen based on their contributions to the improvement of the global fitness.…”

“…Unlike traditional CC, in contribution-based cooperative co-evolution (CBCC) [15], subcomponents are chosen based on their contributions to the improvement of the global fitness. As a result, a subcomponent with a higher contribution to the global fitness will be given more computational resources.…”

“…In addition to the impact that a near-optimal decomposition can have on the performance of CC, it has been shown recently that it is possible to quantify the contribution of a subcomponent to the global fitness [15]. Once this contribution information is calculated, the computational budget can be divided between the subcomponents according to their contributions, unlike traditional CC where the computational budget is equally divided between all subcomponents.…”

“…Once this contribution information is calculated, the computational budget can be divided between the subcomponents according to their contributions, unlike traditional CC where the computational budget is equally divided between all subcomponents. It has been shown that a contribution-based scheme outperforms traditional CC [15].…”

Effective decomposition of large-scale separable continuous functions for cooperative co-evolutionary algorithms

“…A CC framework is appealing, especially for solving large-scale complex problems. Many such techniques have been used for large-scale global optimization [3], [5], [7]- [11], [15]. CC algorithms use a divide-and-conquer approach to decompose a large-scale problem into a set of lower dimensional problems which are easier to optimize.…”