An improved butterfly optimization algorithm for training the feed-forward artificial neural networks

Differential Mutation Incorporated Quantum Honey Badger Algorithm with Dynamic Opposite Learning and Laplace Crossover for Fuzzy Front-End Product Design

In this paper, a multi-strategy fusion enhanced Honey Badger algorithm (EHBA) is proposed to address the problem of easy convergence to local optima and difficulty in achieving fast convergence in the Honey Badger algorithm (HBA). The adoption of a dynamic opposite learning strategy broadens the search area of the population, enhances global search ability, and improves population diversity. In the honey harvesting stage of the honey badger (development), differential mutation strategies are combined, selectively introducing local quantum search strategies that enhance local search capabilities and improve population optimization accuracy, or introducing dynamic Laplacian crossover operators that can improve convergence speed, while reducing the odds of the HBA sinking into local optima. Through comparative experiments with other algorithms on the CEC2017, CEC2020, and CEC2022 test sets, and three engineering examples, EHBA has been verified to have good solving performance. From the comparative analysis of convergence graphs, box plots, and algorithm performance tests, it can be seen that compared with the other eight algorithms, EHBA has better results, significantly improving its optimization ability and convergence speed, and has good application prospects in the field of optimization problems.

Archive-based coronavirus herd immunity algorithm for optimizing weights in neural networks

The success of the supervised learning process for feedforward neural networks, especially multilayer perceptron neural network (MLP), depends on the suitable configuration of its controlling parameters (i.e., weights and biases). Normally, the gradient descent method is used to find the optimal values of weights and biases. The gradient descent method suffers from the local optimal trap and slow convergence. Therefore, stochastic approximation methods such as metaheuristics are invited. Coronavirus herd immunity optimizer (CHIO) is a recent metaheuristic human-based algorithm stemmed from the herd immunity mechanism as a way to treat the spread of the coronavirus pandemic. In this paper, an external archive strategy is proposed and applied to direct the population closer to more promising search regions. The external archive is implemented during the algorithm evolution, and it saves the best solutions to be used later. This enhanced version of CHIO is called ACHIO. The algorithm is utilized in the training process of MLP to find its optimal controlling parameters thus empowering their classification accuracy. The proposed approach is evaluated using 15 classification datasets with classes ranging between 2 to 10. The performance of ACHIO is compared against six well-known swarm intelligence algorithms and the original CHIO in terms of classification accuracy. Interestingly, ACHIO is able to produce accurate results that excel other comparative methods in ten out of the fifteen classification datasets and very competitive results for others.