Using Fitness Dependent Optimizer for Training Multi-layer Perceptron

This study presents a novel training algorithm depending upon the recently proposed Fitness Dependent Optimizer (FDO). The stability of this has been verified and performance-proofed in both exploration exploitation stages using some standard measurements. influenced our target to gauge performance multilayer perceptron neural networks (MLP). combines FDO with MLP (codename FDO-MLP) for optimizing weights biases predict outcomes students. can improve learning system terms educational background students besides increasing their achievements. experimental results approach are affirmed by comparing Back-Propagation (BP) evolutionary models such as cascade (FDO-CMLP), Grey Wolf (GWO) combined (GWO-MLP), modified GWO (MGWO-MLP), (GWO-CMLP), (MGWO-CMLP). qualitative quantitative prove that trainer outperform other approaches different trainers on dataset convergence speed local optima avoidance. FDO-MLP classifies rate 0.97.