2.12.2022: The collide of decision-making, optimization, and machine learning (Aghaei Pour)

In every aspect of our life, from the healthiness of the food we eat versus the taste to how warm our home should be versus the cost or how sustainable we want to be, there are decisions that need to be made. In his thesis, Pouya Aghaei Pour has developed new algorithms that can consider our preferences and assist us in making better decisions.
Moreover, the problems addressed in his thesis are quite complicated. An example addressed in this thesis is finding a balanced configuration between different energy sources in large buildings. Modern buildings can take advantage of different types of fuels. For example, there are fossil fuels, batteries, solar panels, earth warming, and so on. However, how does one decide how much of which energy source one uses in a building? Some of these sources are good for the environment but too expensive, and vice versa. On the other hand, some of these sources may not be very reliable based on the location where we are constructing the building.
Aghaei Pour has used software (provided by the HONDA Research Institute Europe) that simulates the usage of these energy sources and their effect on a building. For example, how much maintenance is needed yearly, how much CO2 is released into the environment, how much initial capital is needed for getting the necessary devices, and so on. Aghaei Pour uses his advanced methods to incorporate the user’s preferences and find tailor-made solutions that the user is interested in. A professional user got to use the method that Aghaei Pour developed in this thesis. Then, the company made software (for internal use) from his research that they are using.
The results of his thesis show how industries can benefit from state-of-the-art research and how the scientific community and industries can work together to provide practical solutions to complex problems.
M.Sc. Pouya Aghaei Pourin laskennallisen tieteen väitöskirjan "Interactive Evolutionary Multiobjective Optimization: Methods and Quality Indicators" tarkastustilaisuus pidetään 2.12.2022 klo 12 alkaen. Vastaväittäjänä Associate Professor LuÃs Paquete (University of Coimbra, Portugali) ja kustoksena professori Kaisa Miettinen (Jyväskylän yliopisto). Väitöstilaisuuden kieli on englanti.
Väitöskirja on julkaistu Jyväskylän yliopiston väitöstutkimusten JYU Dissertations-sarjassa, numero 576, Jyväskylä 2022. ISBN 978-951-39-9233-0 (PDF), URN:ISBN:978-951-39-9233-0, ISSN 2489-9003. Linkki verkkojulkaisuun: .
Yleisö voi seurata väitöstilaisuutta väitössalissa (Agora Auditorio 2) tai verkkovälitteisesti. Linkki suoraan lähetykseen: