ABSTRACT
A finite set of best trade-off solutions, or Pareto frontier, is the searching result of a multi-objective optimization algorithm against a multi-objective optimization problem. However, not all the solutions of the set are equally important to decision makers. They typically utilize only a few outstanding solutions, and the solutions located at the maximum convex bulge on the Pareto frontier are usually embraced when no preference is available; they are called knee solutions. There are several knee searching algorithms in the last decades, but most of them failed to isolate the knee solutions from the near knee solutions. In this paper, we propose a posteriori knee searching algorithm that can identify and isolate the knee solutions, based on the farthest distance to a hyperplane among the neighborhood solutions. The proposed algorithm is tested against well-known benchmark problems: ZDT3, DEB2DK and DEB3DK. The results show that the proposed algorithm can identify outstanding solutions which are knee solutions accurately.
- Deb, K., Pratap, A., AgarwaL, S., and Meyarivan, T. 2002. A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans on Evol Comp 6, 2, 182--197.Google ScholarDigital Library
- Qingfu, Z. and HUI, L. 2007. MOEA/D: A multiobjective evolutionary algorithm based on decomposition. IEEE Trans Evol Comp 11, 6, 712--731.Google ScholarDigital Library
- Branke, J., Kaußbler, T., and Schmeck H. 2001. Guidance in evolutionary multi-objective optimization. Advances in Eng. Soft. 32, 6 (2001/06/01/), 499--507.Google Scholar
- Wang, R., Purshouse, R.C., and Fleming, P.J. 2013. Preference-inspired coevolutionary algorithms for many-objective optimization. IEEE Trans on Evoly Comp 17, 4 (Aug), 474--494.Google Scholar
- Cheng, R., Jin, Y.C., Olhofer, M., and Sendhoff, B. 2016. A reference vector guided evolutionary algorithm for many-objective optimization. IEEE Trans on Evol Comp 20, 5 (Oct), 773--791.Google ScholarCross Ref
- Das, I. 1999. On characterizing the "knee" of the Pareto curve based on Normal-Boundary Intersection. Structural and Multidisciplinary Optimization 18, 2-3, 107--115.Google ScholarCross Ref
- Zhang, X., Tian, Y., and Jin, Y. 2015. A knee point-driven evolutionary algorithm for many-objective optimization. IEEE Transs on Evol Comp 19, 6, 761--776.Google ScholarCross Ref
- Sudeng, S. and Wattanapongsakorn, N. 2016. A decomposition-based approach for knee solution approximation in multi-objective optimization. In Proc. of the 2016 Congress on Evol Comp(IEEE), Vancouver, Cananda, 3710--3717.Google Scholar
- Zou, J., et al. 2019. A knee-point-based evolutionary algorithm using weighted subpopulation for many-objective optimization. Swarm and Evol Comp 47(Jun), 33--43.Google ScholarCross Ref
- Dhiman, G. and Kumar, V., 2019. KnRVEA: A hybrid evolutionary algorithm based on knee points and reference vector adaptation strategies for many-objective optimization. Applied Intelligence 49, 7 (Jul), 2434--2460.Google ScholarDigital Library
- Branke, E., et al. 2004. Finding knees in multi-objective optimization. Parallel Problem Solving from Nature - Ppsn Viii 3242, 722--731.Google Scholar
- Deb, K. and Gupta, S. 2011. Understanding knee points in bicriteria problems and their implications as preferred solution principles. Eng Opt 43, 11 (2011/11/01), 1175--1204.Google Scholar
- Sudeng, S. and Wattanapongsakorn, N. 2015. Post paretooptimal pruning algorithm for multiple objective optimization using specific extended angle dominance. Eng Apps of AI 38, 221--236.Google ScholarCross Ref
- Ramirez-Atencia, C., Mostaghim, S., and Camacho, D. 2017. A knee point based evolutionary multi-objective optimization for mission planning problems. In Proceedings of the Genetic and Evol Comp Conference(Berlin, Germany2017), ACM, 3071319, 1216--1223.Google ScholarDigital Library
- Braun, M., Shukla, P., and Schmech, H. 2017. Angle-based preference models in multi-objective optimization. Lecture Notes in Comp Sci., 88--102.Google Scholar
- Ye, X., et al. 2017. User-oriented many-objective cloud workflow scheduling based on an improved knee point driven evolutionary algorithm. Knowledge-Based Systems 135(Nov), 113--124.Google ScholarDigital Library
- Rachmawati, L. and Srinivasan, D. 2009. Multiobjective evolutionary algorithm with controllable focus on the knees of the pareto front. IEEE Trans on Evol Compn 13, 4, 810--824.Google ScholarDigital Library
- Bhattacharjee, K.S., et al. 2017. Bridging the gap: Many-objective optimization and informed decision-making. IEEE Trans on Evol Comp 21, 5, 813--820.Google ScholarCross Ref
- Yu, G., Jin, Y., and Olhofer, M. 2018. A method for a posteriori identification of knee points based on solution density. In 2018 IEEE World Congress on Computational Intelligence (WCCI 2018), Rio de Janeiro, Brazil, 1--8.Google Scholar
Index Terms
- Finding Outstanding Solutions for Multi-objective Optimization Problems
Recommendations
Multi-Objective Optimisation Problems: A Symbolic Algorithm for Performance Measurement of Evolutionary Computing Techniques
EMO '09: Proceedings of the 5th International Conference on Evolutionary Multi-Criterion OptimizationIn this paper, a symbolic algorithm for solving constrained multi-objective optimisation problems is proposed. It is used to get the Pareto optimal solutions as functions of KKT multipliers $\overrightarrow{\lambda}$ for multi-objective problems with ...
Meta-level multi-objective formulations of set optimization for multi-objective optimization problems: multi-reference point approach to hypervolume maximization
GECCO Comp '14: Proceedings of the Companion Publication of the 2014 Annual Conference on Genetic and Evolutionary ComputationHypervolume has been frequently used as an indicator to evaluate a solution set in indicator-based evolutionary algorithms (IBEAs). One important issue in such an IBEA is the choice of a reference point. A different solution set is often obtained from a ...
Finding multiple solutions for multimodal optimization problems using a multi-objective evolutionary approach
GECCO '10: Proceedings of the 12th annual conference on Genetic and evolutionary computationIn a multimodal optimization task, the main purpose is to find multiple optimal (global and local) solutions associated with a single objective function. Starting with the preselection method suggested in 1970, most of the existing evolutionary ...
Comments