Abstract

Nature-inspired optimization techniques are popular in the case of non-convex and derivative-free optimization problems. The aim of the project is to study and implement a hybrid algorithm and provide some experiments using freely available datasets. The optimization task here would be hyperparameter tuning of ML techniques. In this report, an effective combination of 10 swarm intelligence algorithms, namely (IWO),(CS), (GA), (PSO), (Firefly), (GWO), (ABC), (DE), (GSO), (FSS) has been proposed. This hybridization called JSI consists of two main phases of running the algorithms in parallel and in each iteration, forming a new population from the result of each algorithm. the experi- ment was performed on one benchmark function.