A New Paradigm in Interactive Evolutionary Multiobjective Optimization
| Autor: | Jussi Hakanen, Kaisa Miettinen, Bhupinder Singh Saini |
|---|---|
| Přispěvatelé: | Bäck, Thomas, Preuss, Mike, Deutz, André, Wang, Hao, Doerr, Carola, Emmerich, Michael, Trautmann, Heike |
| Rok vydání: | 2020 |
| Předmět: |
050101 languages & linguistics
Mathematical optimization Computer science media_common.quotation_subject decision maker Evolutionary algorithm päätöksentukijärjestelmät evoluutiolaskenta preference information 02 engineering and technology Space (commercial competition) Multi-objective optimization optimointi achievement scalarizing functions algoritmit 0202 electrical engineering electronic engineering information engineering 0501 psychology and cognitive sciences Quality (business) evolutionary algorithms Function (engineering) media_common business.industry 05 social sciences interactive methods Modular design Decision maker monitavoiteoptimointi Preference 020201 artificial intelligence & image processing business |
| Zdroj: | Parallel Problem Solving from Nature – PPSN XVI ISBN: 9783030581145 PPSN (2) |
| DOI: | 10.1007/978-3-030-58115-2_17 |
| Popis: | Over the years, scalarization functions have been used to solve multiobjective optimization problems by converting them to one or more single objective optimization problem(s). This study proposes a novel idea of solving multiobjective optimization problems in an interactive manner by using multiple scalarization functions to map vectors in the objective space to a new, so-called preference incorporated space (PIS). In this way, the original problem is converted into a new multiobjective optimization problem with typically fewer objectives in the PIS. This mapping enables a modular incorporation of decision maker’s preferences to convert any evolutionary algorithm to an interactive one, where preference information is directing the solution process. Advantages of optimizing in this new space are discussed and the idea is demonstrated with two interactive evolutionary algorithms: IOPIS/RVEA and IOPIS/NSGA-III. According to the experiments conducted, the new algorithms provide solutions that are better in quality as compared to those of state-of-the-art evolutionary algorithms and their variants where preference information is incorporated in the original objective space. Furthermore, the promising results require fewer function evaluations. peerReviewed |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|