A multipopulation evolutionary framework with Steffensen’s method for dynamic multiobjective optimization problems

Dynamic multiobjective optimization problems (DMOPs) require the evolutionary algorithms that can track the moving Pareto-optimal fronts efficiently. This paper presents a dynamic multiobjective evolutionary framework (DMOEF-MS), which adopts a novel multipopulation structure and Steffensen’s method to solve DMOPs. In DMOEF-MS, only one population deals with the original DMOP, while the others focus on single-objective problems that are generated by the weighted summation of the original DMOP. Then, Steffensen’s method is used to control the evolving process in two ways: prediction and diversity-maintenance. Particularly, the prediction strategy is devised to predict the next promising positions for the individuals that handle single-objective problems, and the diversity-maintenance strategy is used to increase population diversity before the environment changes and reinitialize the multiple populations after the environment changes. This paper gives a comprehensive comparison of DMOEF-MS with some state-of-the-art DMOEAs on 14 DMOPs and the experimental results demonstrate the effectiveness of the proposed algorithm.

Modelling other agents through evolutionary behaviours

Modelling other agents is a challenging topic in artificial intelligence research particularly when a subject agent needs to optimise its own decisions by predicting their behaviours under uncertainty. Existing research often leads to a monotonic set of behaviours for other agents so that a subject agent can not cope with unexpected decisions from the other agents. It requires creative ideas about developing diversity of behaviours so as to improve the subject agent’s decision quality. In this paper, we resort to evolutionary computation approaches to generate a new set of behaviours for other agents and solve the complicated agents’ behaviour search and evaluation issues. The new approach starts with the initial behaviours that are ascribed to the other agents and expands the behaviours by using a number of genetic operators in the behaviour evolution. This is the first time that evolutionary techniques are used to modelling other agents in a general multiagent decision framework. We examine the new methods in two well-studied problem domains and provide experimental results in support.