Seminarium: pyMCMA: Uniformly distributed Pareto-front representation

Zapraszamy na wspólne seminarium Katedry Informatyki i Inżynierii Systemów na Wydziale Informatyki i Zarządzania Politechniki Wrocławskiej, Centrum Technik Informatycznych WIT, Zakładu Wspomagania Decyzji w Warunkach Ryzyka IBS PAN

Podczas spotkania Marek Makowski, Janusz Granat, Andrii Shekhovtsov, Zbigniew Nahorski i Jinyang Zhao przedstawią referat pt. pyMCMA: Uniformly distributed Pareto-front representation.

Abstract
Multiple-Criteria Model Analysis (MCMA) offers scientific support in model-based problem-solving through comprehensive analysis of trade-offs between simultaneously attainable goals for often conflicting objectives. Scientific support should be objective. Therefore, scientific analyses should be preference-free and uniformly cover the whole set of Pareto solutions. The presentation will introduce the pyMCMA package that seamlessly integrates the MCMA with independently developed substantive models, and autonomously computes Pareto-front representation composed of efficient solutions that are uniformly distributed in terms of distances between neighbor solutions. The results are illustrated by plots, also for more than three criteria. Moreover, the pyMCMA exports the Pareto-front representation for in-depth analysis in the criteria space, as well as optionally exports the results for problems-specific analysis in the substantive model's variables space.

Seminarium odbędzie się 28 lutego 2024 (środa) o godzinie 13:15 za pośrednictwem platformy Skype: link do spotkania.