Multi-Guide Particle Swarm Optimization for Large-Scale Multi-Objective Optimization Problems
Abstract
Multi-guide particle swarm optimization (MGPSO) is a novel metaheuristic for multi-objective optimization based on particle swarm optimization (PSO). MGPSO has been shown to be competitive when compared with other state-of-the-art multi-objective optimization algorithms for low-dimensional problems. However, to the best of the author’s knowledge, the suitability of MGPSO for high-dimensional multi-objective optimization problems has not been studied. One goal of this thesis is to provide a scalability study of MGPSO in order to evaluate its efficacy for high-dimensional multi-objective optimization problems. It is observed that while MGPSO has comparable performance to state-of-the-art multi-objective optimization algorithms, it experiences a performance drop with the increase in the problem dimensionality. Therefore, a main contribution of this work is a new scalable MGPSO-based algorithm, termed cooperative co-evolutionary multi-guide particle swarm optimization (CCMGPSO), that incorporates ideas from cooperative PSOs. A detailed empirical study on well-known benchmark problems comparing the proposed improved approach with various state-of-the-art multi-objective optimization algorithms is done. Results show that the proposed CCMGPSO is highly competitive for high-dimensional problems.Collections
Related items
Showing items related by title, author, creator and subject.
-
Comparative Study On Cooperative Particle Swarm Optimization Decomposition Methods for Large-scale OptimizationThe vast majority of real-world optimization problems can be put into the class of large-scale global optimization (LSOP). Over the past few years, an abundance of cooperative coevolutionary (CC) algorithms has been proposed to combat the challenges of LSOP’s. When CC algorithms attempt to address large scale problems, the effects of interconnected variables, known as variable dependencies, causes extreme performance degradation. Literature has extensively reviewed approaches to decomposing problems with variable dependencies connected during optimization, many times with a wide range of base optimizers used. In this thesis, we use the cooperative particle swarm optimization (CPSO) algorithm as the base optimizer and perform an extensive scalability study with a range of decomposition methods to determine ideal divide-and-conquer approaches when using a CPSO. Experimental results demonstrate that a variety of dynamic regrouping of variables, seen in the merging CPSO (MCPSO) and decomposition CPSO (DCPSO), as well varying total fitness evaluations per dimension, resulted in high-quality solutions when compared to six state-of-the-art decomposition approaches.
-
Implementing the OPTIMAL model : the impact on students' motivation in an elementary school games environmentOptimal challenge occurs when an individual perceives the challenge of the task to be equaled or matched by his or her own skill level (Csikszentmihalyi, 1990). The purpose of this study was to test the impact of the OPTIMAL model on physical education students' motivation and perceptions of optimal challenge across four games categories (i. e. target, batting/fielding, net/wall, invasion). Enjoyment, competence, student goal orientation and activity level were examined in relation to the OPTIMAL model. A total of 22 (17 M; 5 F) students and their parents provided informed consent to take part in the study and were taught four OPTIMAL lessons and four non-OPTIMAL lessons ranging across the four different games categories by their own teacher. All students completed the Task and Ego in Sport Questionnaire (TEOSQ; Duda & Whitehead, 1998), the Intrinsic Motivation Inventory (IMI; McAuley, Duncan, & Tanmien, 1987) and the Children's Perception of Optimal Challenge Instrument (CPOCI; Mandigo, 2001). Sixteen students (two each lesson) were observed by using the System for Observing Fitness Instruction Time tool (SOFTT; McKenzie, 2002). As well, they participated in a structured interview which took place after each lesson was completed. Quantitative results concluded that no overall significant difference was found in motivational outcomes when comparing OPTIMAL and non-OPTIMAL lessons. However, when the lessons were broken down into games categories, significant differences emerged. Levels of perceived competence were found to be higher in non-OPTIMAL batting/fielding lessons compared to OPTIMAL lessons, whereas levels of enjoyment and perceived competence were found to be higher in OPTIMAL invasion lessons in comparison to non-OPTIMAL invasion lessons. Qualitative results revealed significance in feehngs of skill/challenge balance, enjoyment and competence in the OPTIMAL lessons. Moreover, a significance of practically twice the active movement time percentage was found in OPTIMAL lessons in comparison to non-OPTIMAL lessons.
-
Characterizing Dynamic Optimization Benchmarks for the Comparison of Multi-Modal Tracking AlgorithmsPopulation-based metaheuristics, such as particle swarm optimization (PSO), have been employed to solve many real-world optimization problems. Although it is of- ten sufficient to find a single solution to these problems, there does exist those cases where identifying multiple, diverse solutions can be beneficial or even required. Some of these problems are further complicated by a change in their objective function over time. This type of optimization is referred to as dynamic, multi-modal optimization. Algorithms which exploit multiple optima in a search space are identified as niching algorithms. Although numerous dynamic, niching algorithms have been developed, their performance is often measured solely on their ability to find a single, global optimum. Furthermore, the comparisons often use synthetic benchmarks whose landscape characteristics are generally limited and unknown. This thesis provides a landscape analysis of the dynamic benchmark functions commonly developed for multi-modal optimization. The benchmark analysis results reveal that the mechanisms responsible for dynamism in the current dynamic bench- marks do not significantly affect landscape features, thus suggesting a lack of representation for problems whose landscape features vary over time. This analysis is used in a comparison of current niching algorithms to identify the effects that specific landscape features have on niching performance. Two performance metrics are proposed to measure both the scalability and accuracy of the niching algorithms. The algorithm comparison results demonstrate the algorithms best suited for a variety of dynamic environments. This comparison also examines each of the algorithms in terms of their niching behaviours and analyzing the range and trade-off between scalability and accuracy when tuning the algorithms respective parameters. These results contribute to the understanding of current niching techniques as well as the problem features that ultimately dictate their success.
