Search

This paper presents a multi-objective version of the artificial vultures optimization algorithm (AVOA) for a multi-objective optimization problem called a multi-objective AVOA (MOAVOA). The inspirational concept of the AVOA is based on African vultures' lifestyles. Archive, grid, and leader selection mechanisms are used for developing the MOAVOA. The proposed MOAVOA algorithm is tested oneight real-world engineering design problems and seventeen unconstrained and constrained mathematical optimization problems to investigates its appropriateness in estimating Pareto optimal solutions. Multi-objective particle swarm optimization, multi-objective ant lion optimization, multi-objective multi-verse optimization, multi-objective genetic algorithms, multi-objective salp swarm algorithm, and multi-objective grey wolf optimizer are compared with MOAVOA using generational distance, inverted generational distance, maximum spread, and spacing performance indicators. This paper demonstrates that MOAVOA is capable of outranking the other approaches. It is concluded that the proposed MOAVOA has merits in solving challenging multi-objective problems. [ABSTRACT FROM AUTHOR]

One of the useful knowledge representation tools, which can describe the joint probability distribution between some random variables with a graphical model and can be trained by a dataset, is the Bayesian network (BN). A BN is composed of a network structure and a conditional probability distribution table for each node. Discovering an optimal BN structure is an NP‐hard optimization problem that various meta‐heuristic algorithms are applied to solve this problem by researchers. The genetic algorithms, ant colony optimization, evolutionary programming, artificial bee colony, and bacterial foraging optimization are some of the meta‐heuristic methods to solve this problem using a dataset. Most of these methods are applying a scoring metric to generate the best network structure from a set of candidates. A Fixed Structure Learning Automata‐Based (FSLA‐B) algorithm is presented in this paper to solve the structure learning problem of BNs. There is a fixed structure learning automaton for each pair of vertices in the BN's graph structure in the proposed algorithm. The action of this automaton determines the presence and direction of an edge between the vertices. The proposed algorithm performs a guided search procedure using the FSLA and escapes from local optimums. Several datasets are utilised in this paper to evaluate the performance of the proposed algorithm. By performing various experiments, multiple meta‐heuristic algorithms are compared with the introduced new one. The obtained results represented that the proposed algorithm could produce competitive results and find the near‐optimal solution for the BN structure learning problem. [ABSTRACT FROM AUTHOR]

• The population is divided into three sub-populations. • A novel population evolution strategy is introduced to improve the basic WOA. • The proposed algorithm (MEWOA) has high solution accuracy and converges fast. • CEC 2019 test suite and 30 benchmark functions are used to test MEWOA. • MEWOA is applied to four engineering design problems. The whale optimization algorithm (WOA) tends to suffer from slow convergence speed and quickly falling into the local optimum. In this work, a WOA variant is proposed based on multi-population evolution (MEWOA) to address these problems. Firstly, individuals are classified into three equal-sized sub-populations: exploratory sub-population, exploitative sub-population, and modest sub-population, according to their fitness. Secondly, the moving strategies of each sub-population are assigned using different mechanisms. The exploratory and exploitative sub-populations perform global and local search, respectively, while the modest sub-population randomly explores or exploits the search space. Finally, we introduce a novel population evolution strategy to help MEWOA improve its global optimization ability and avoid local optimum. MEWOA is compared with five state-of-the-art WOA variants and seven basic metaheuristic algorithms over 30 benchmark functions with dimensions of 100, 500, 1000, and 2000 respectively. It is observed that MEWOA achieves faster convergence speed, shows shorter runtime, and provides higher solution accuracy than other algorithms on the majority of benchmark functions. In addition, we tested MEWOA's ability to solve challenging real-world and constrained optimization problems on the CEC 2019 test suite and four engineering design problems. The experimental results demonstrate the competitiveness and merits of the proposed MEWOA algorithm. [ABSTRACT FROM AUTHOR]