Teaching-learning-based optimization with two-stage initialization

The teaching-learning-based optimization (TLBO) algorithm is an evolutionary algorithm based on simulated teaching-learning procedure in a class. This algorithm requires no controlling parameters and is moulded based on the effect of influence of teacher on the output of learners and hence can be implemented easily and also requires less computational memory compared to other evolutionary algorithms. This paper presents the (TLBO) algorithm with two-stage initialization (TSI). The TLBO with TSI (TLBO-TSI) is tested with the Rastrigin benchmark function and then used to minimize least-squares (LS), Huber (HU) and Hampel (HA) cost functions. Simulation results are presented to show the evolutionary behavior of the TLBO with TSI. Simulation results show that the TLBO-TSI performs better compared to existing TLBO.