Research indexJournal article
An Indicator Based Multiobjective Evolutionary Algorithm Utilizing Inverse Modeling
Zheng, Liu, Tan and 1 other
28 Jul 2024
NobleIDark:/48914/w1/20260525/25C5DB1FDOI10.23919/ccc63176.2024.10661744
An Indicator Based Multiobjective Evolutionary Algorithm Utilizing Inverse Modeling
Zheng, Liu, Tan and 1 other · 2024 · Subscription
NobleIDark:/48914/w1/20260525/25C5DB1FDOI10.23919/ccc63176.2024.10661744
The full text is at the publisher
You can read the abstract here, or open it at the publisher.
Abstract
In recent years, inverse models have attracted an increasing interest in the optimization field due to their superior ability on the multiobjective optimization. Specially, inverse models can commonly be employed to approximate Pareto optimal set by sampling Pareto front, which achieve the mapping from the objective space to the decision space. In this work, we design an indicator based evolutionary algorithm (IM-IBEA), equipping with an inverse model for the multiobjective optimization. IMIBEA is evaluated on 27 WFG test problems with 30-300 decision variables, which achieves best performance in 22 out of 27 test instances. The experimental results indicate that IM-IBEA has superior performance compared than the state-of-the-art methods.