Multiobjective multifactorial immune algorithm for multiobjective multitask optimization problems

Abstract

Inspired by human brains’ ability to solve multiple tasks simultaneously, evolutionary multitasking is proposed to improve the overall efficiency of optimizing multiple tasks simultaneously by reusing the learned knowledge. The immune algorithm is inspired by the biological immune system that has been proven to be effective in many practical multiobjective optimization problems, with efficient convergence and search efficiency. In this paper, a novel multiobjective multifactorial immune algorithm is proposed with a novel information transfer method to solve multiobjective multitask optimization problems. For each task, information advantageous for this task will be transferred from the others to accelerate convergence through the proposed information transfer method. Finally, the proposed algorithm is compared with the state-of-the-art multiobjective evolutionary multitasking algorithms and the classic multiobjective evolutionary algorithms. The experimental results on the classical multiobjective multitask and the multiobjective many-task test suites demonstrate that the proposed algorithm provides very promising performances.