Research On Meta-heuristic Algorithms For Solving Multi-objective Unconstrained Binary Quadratic Programming Problems

Multi-objective optimization,as an important branch of operational optimization,focuses on the simultaneous optimization of multiple objective values.It is widely used in computer science,artificial intelligence,financial analysis,and other disciplines and engineering fields.Many problems in the real world can be solved by using combinatorial optimization methods,these problems often have conflicting goals that need to be optimized simultaneously.In order to achieve the best optimization results,they often require trade-offs between goals.The unconstrained binary quadratic programming problem,as a class of classical combinatorial optimization problems,has always been a hot topic of research.The academic community has proposed exact algorithms,heuristic algorithms and meta-heuristic algorithms.Due to the NP-hard nature of the problem,the exact algorithms often fail to obtain a satisfactory solution in acceptable time.In recent years,many intelligent optimization methods have been proposed,such as genetic algorithm,tabu search algorithm,simulated annealing algorithm,ant colony optimization algorithm etc.These optimization algorithms pay more attention to the speed and efficiency of the calculation,and do not seek to obtain the optimal solution.In this thesis,based on the hyper-volume contribution selection algorithm framework,combined with genetic algorithm and hybrid perturbation strategy,three kinds of metaheuristic algorithms are proposed.The hyper-volume contribution selection algorithm is used to allocate the fitness value for individuals.The genetic algorithm uses crossover operators and mutation operations to produce high quality and diverse offsprings.The hybrid perturbation strategy is used to expand the search space of the algorithm,enhance the search ability of the algorithm,and then obtain a diversity solution set.Experimental tests on the standard examples show that the proposed algorithms are very effective in comparison with iterative local search algorithm,indicator based multi-objective local search and hyper-volume based multi-objective local search.