ABSTRACT
Constraints are a reality in many real-world application areas, adding feasibility to the existing multiple objective challenge. Further, the presence of complex constraints poses a significant challenge to multi-objective evolutionary algorithms. A recently proposed biphasic multi-objective evolutionary framework for constrained multi-objective optimization problems is the Push and Pull Search framework. This framework benefits from a strong exploration of the constrained landscape during the search for the unconstrained Pareto-Front during the first phase. The work herein extends the Push and Pull Search framework, extending landscape information gathering in the push phase; adding a binary search of the feasible and infeasible regions and creating a suitably diverse population and improved initialization for the push phase.
- Zhun Fan, Wenji Li, Xinye Cai, Han Huang, Yi Fang, Yugen You, Jiajie Mo, Caimin Wei, and Erik Goodman. 2019. An improved epsilon constraint-handling method in MOEA/D for CMOPs with large infeasible regions. Soft Computing 23, 23 (2019), 12491--12510.Google ScholarCross Ref
- Zhun Fan, Wenji Li, Xinye Cai, Hui Li, Caimin Wei, Qingfu Zhang, Kalyanmoy Deb, and Erik Goodman. 2019. Push and pull search for solving constrained multi-objective optimization problems. Swarm and evolutionary computation 44 (2019), 665--679.Google Scholar
- Stephanus Daniel Handoko, Chee Keong Kwoh, and Yew-Soon Ong. 2010. Feasibility structure modeling: An effective chaperone for constrained memetic algorithms. IEEE Transactions on Evolutionary Computation 14, 5 (2010), 740--758.Google ScholarDigital Library
- Guillermo Leguizamon and Carlos A Coello Coello. 2008. Boundary search for constrained numerical optimization problems with an algorithm inspired by the ant colony metaphor. IEEE transactions on evolutionary computation 13, 2 (2008), 350--368.Google Scholar
- Guillermo Leguizamón and Carlos Coello Coello. 2009. Boundary Search for Constrained Numerical Optimization Problems. Springer Berlin Heidelberg, Berlin, Heidelberg, 25--49. Google ScholarCross Ref
- Hai-Lin Liu, Chaoda Peng, Fangqing Gu, and Jiechang Wen. 2016. A constrained multi-objective evolutionary algorithm based on boundary search and archive. International Journal of Pattern Recognition and Artificial Intelligence 30, 01 (2016), 1659002.Google ScholarCross Ref
- Haiping Ma, Haoyu Wei, Ye Tian, Ran Cheng, and Xingyi Zhang. 2021. A multistage evolutionary algorithm for multi-objective optimization with complex constraints. Information Sciences 560 (2021), 68--91.Google ScholarCross Ref
- Zhongwei Ma and Yong Wang. 2019. Evolutionary constrained multiobjective optimization: Test suite construction and performance comparisons. IEEE Transactions on Evolutionary Computation 23, 6 (2019), 972--986.Google ScholarDigital Library
- Karam M Sallam, Ruhul A Sarker, and Daryl L Essam. 2017. Reduced search space mechanism for solving constrained optimization problems. Engineering Applications of Artificial Intelligence 65 (2017), 147--158.Google ScholarDigital Library
- Qingfu Zhang and Hui Li. 2007. MOEA/D: A multiobjective evolutionary algorithm based on decomposition. IEEE Transactions on evolutionary computation 11, 6 (2007), 712--731.Google ScholarDigital Library
Index Terms
- Extending the push and pull search framework with boundary search for constrained multi-objective optimization
Recommendations
A dynamic dual-population co-evolution multi-objective evolutionary algorithm for constrained multi-objective optimization problems
AbstractMany multi-objective evolutionary algorithms are proposed to handle constrained multi-objective optimization problems. Nevertheless, they often fail to appropriately balance feasibility, convergence and diversity of the population. ...
Highlights- A dual-population MOEA for CMOPs.
- Dynamic dual-population co-evolution.
Binary search based boundary elimination selection in many-objective evolutionary optimization
Display Omitted This paper proposed a novel selection strategy the boundary elimination selection based on the binary search.This paper uses the binary search during the environmental selection to adjust the value to keep the stability of the population ...
Search space-based multi-objective optimization evolutionary algorithm
Evolutionary multi-objective optimization (EMO) algorithms are actively used for answering optimization problems with multiple contradictory objectives and scheming interpretable and precise real-time applications. A majority of existing EMO algorithms ...
Comments