Paper Info
Title
Collective preferences in evolutionary multi-objective optimization: techniques and potential contributions of collective intelligence
Abstract
This paper reviews suitable techniques of interactive and preference-based evolutionary multi-objective algorithms to achieve feasible solutions in Pareto-optimal front. We discuss about possible advantages of collective environments to aggregate consistent preferences in the optimization process. Decision maker can highlight the regions of Pareto frontier that are more relevant to him and focus the search only on those areas previously selected. In addition, interactive and cooperative genetic algorithms work on refining users' preferences throughout the optimization process to improve the reference point or fitness function. Nevertheless, expressing preferences from a unique or small group of decision makers may raise unilateral choices issues and pour hints in terms of search parameter. Supported by a large group of human interaction, collective intelligence is suggested to enhance multi-objective results and explore a wider variety of answers.
Year
DOI
Venue
2015
10.1145/2695664.2695926
SAC 2015: Symposium on Applied Computing Salamanca Spain April, 2015
Keywords
Field
DocType
collective intelligence, preferences, reference points, evolutionary multi-objective optimization algorithms
Computer science,Collective intelligence,Multi-objective optimization,Human interaction,Fitness function,Decision maker,Pareto principle,Genetic algorithm,Management science
Conference
ISBN
Citations 
PageRank 
978-1-4503-3196-8
0
0.34
References 
Authors
14
4
Name
Order
Citations
PageRank
Daniel Cinalli100.68
Luis Martí2439.51
Nayat Sánchez Pi34815.93
Ana Cristina Bicharra Garcia424750.45