Data-driven Interactive Multiobjective Optimization: Challenges and a Generic Multi-agent Architecture
Year: 2020
Type: conference-paper
Abstract: In many decision making problems, a decision maker needs computer support in finding a good compromise between multiple conflicting objectives that need to be optimized simultaneously. Interactive multiobjective optimization methods have a lot of potential for solving such problems. However, the growth of complexity in problem formulations and the abundance of data bring new challenges to be addressed by decision makers and method developers. On the other hand, advances in the field of artificial intelligence provide opportunities in this respect. We identify challenges and propose directions of addressing them in interactive multiobjective optimization methods with the help of multiple intelligent agents. We describe a generic architecture of enhancing interactive methods with specialized agents to enable more efficient and reliable solution processes and better support for decision makers.
Source: Procedia Computer Science
Authors: Bekir Afsar, Dmitry Podkopaev, Kaisa Miettinen
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
Funder: Academy of Finland
Awards: G4797184450, G8782467213
OpenAlex ID: W3090818718
