Abstract
Teaching-learning-based optimization (TLBO) algorithm has been shown to be an effective optimization algorithm. However, it is easily trapped into local optima when the global optimal solution of the function to be optimized is at the original dot or around the original dot. This paper presents a novel TLBO variant by incorporating multiobjective sorting-based mechanism and cooperative learning strategy to alleviate this problem. Taking advantages of multiobjective optimization in maintaining good population diversity, several teachers are selected based on non-dominated sorting, so as to guide learners to learn more effectively. In addition, the proposed algorithm adopts cooperative learning, including learning within and between groups, to improve the search ability of the algorithm. Experimental and statistical analyses are performed on CEC2014 benchmark functions. The experimental results demonstrate the effectiveness of the proposed algorithm in comparison with other variants of TLBO and other state-of-the-art optimization algorithms.
Highlights
The optimization problem has long been an interest research topic due to its wide application in real-world scenarios, including large-scale optimization, multiobjective optimization, big data, feature selection, and so on [1]–[4]
Inspired by natural and physical phenomena, evolutionary algorithms (EAs) are booming because nature-inspired mechanisms can be transformed into search mechanisms, which can be effectively implemented to improve the solution search ability for optimization problem [5]
In view of limitations of Teaching-Learning-Based Optimization (TLBO), this paper proposes a novel TLBO with multiobjective sorting-based and cooperative learning (MSCTLBO)
Summary
The optimization problem has long been an interest research topic due to its wide application in real-world scenarios, including large-scale optimization, multiobjective optimization, big data, feature selection, and so on [1]–[4]. The teaching control parameter is dynamically adjusted according to fitness values of learners, while the learning control parameter is adaptively adjusted according to the current generation and the maximum generation This mechanism can effectively improve the exploration ability and the exploitation ability of the algorithm. A self-adaptive strategy is employed to justify the population sizes for the teaching and learning reproduction. Chen et al [30] proposed a variant of TLBO with multi-classes cooperation and simulated annealing operator (SAMCCTLBO) to improve the learning ability of learners and the diversity of the whole class. The proposed algorithm makes the best of the excellent global optimization ability of NNA and the fast convergence rate of TLBO by dynamic grouping mechanism. The value of TF can be either 1 or 2
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
