2820. DESDEO 2.0: the new and improved open-source framework for interactive multiobjective optimization
Invited abstract in session WA-51: Advances in multiobjective optimization software, stream Multiobjective and vector optimization.
Wednesday, 8:30-10:00Room: Parkinson B22
Authors (first author is the speaker)
| 1. | Giovanni Misitano
|
| P.O. Box 35, University of Jyväskylä | |
| 2. | Bhupinder Singh Saini
|
| University of Jyväskylä | |
| 3. | Juho Roponen
|
| Faculty of Information Technology, University of Jyvaskyla | |
| 4. | Juuso Pajasmaa
|
| 5. | Giomara Lárraga
|
| University of Jyvaskyla | |
| 6. | Babooshka Shavazipour
|
| University of Jyvaskyla | |
| 7. | Kaisa Miettinen
|
| University of Jyväskyla |
Abstract
DESDEO is an open-source Python framework for implementing and applying interactive multiobjective optimization methods. Interactive methods involve the decision maker in the optimization process, allowing them to iteratively provide preferences and explore available solutions. Preferences can be utilized to find the most promising solutions for the decision maker, helping them in tackling the tradeoffs among the conflicting objective functions of the problem being solved.
DESDEO 2.0 is the newest major release of the DESDEO framework. The development of new framework has consisted of a full restructuring. In this process, we have developed a novel way to model and represent a vast array of different multiobjective optimization problem types, including non-linear and data-driven problems. Once a problem is modeled, it can be solved using a variety of interactive methods, ranging from exact mathematical methods to heuristic-based evolutionary optimization. This makes DESDEO applicable to a wide range of domains.
This talk will give an overview of DESDEO 2.0’s newest features and how to get started using the framework. The framework is aimed to both practitioners and researchers in multiobjective optimization. DESDEO is very modular, making it suitable for experimenting with new ideas, and allowing the reuse of existing components. Our goal is to make DESDEO a collaborative open-source hub for researching and applying interactive multiobjective optimization.
Keywords
- Multi-Objective Decision Making
- Programming, Multi-Objective
- Software
Status: accepted
Back to the list of papers