Teaching-Learning Based Optimization Algorithm

777 WordsFeb 20, 20183 Pages
Teaching−Learning-Based Optimization Algorithm TLBO is a recently proposed meta-heuristic that imitates a successful and dynamic educational strategy in a classroom [29-31]. Similar to most evolutionary algorithms, TLBO is a population-based algorithm. The population consists of some students and a teacher. The teacher is the most knowledgeable one in the population. The main advantage of this algorithm over other evolutionary algorithms is that TLBO has no adjustable parameters, so there is no need to design a tuning mechanism for the parameters. The educational strategy of this algorithm includes both direct and interactive instruction. Actually, not only can the students be affected by their teacher, but also they can affect each other. TLBO algorithm can be divided into two phases: 1- Teaching phase (direct instruction) 2- Learning phase (interactive instruction). In the teaching phase, the teacher provides information for all the students and the students learn from their teacher, while in the learning phase they can learn from each other and develop their skills. The pseudo code of the proposed TLBO algorithm is shown in Figure 1. Teaching phase The best member is selected as the teacher in each iteration. The teacher trains his/her students. In practice the teacher can only improve the mean of the students’ knowledge. Students’ improvement depends on the students’ aptitude for learning. The knowledge of each student is changed according to the following equations:
Open Document