Show simple item record

dc.contributor.authorMazumdar, Atanu
dc.date.accessioned2021-12-03T11:09:36Z
dc.date.available2021-12-03T11:09:36Z
dc.date.issued2021
dc.identifier.isbn978-951-39-8919-4
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/78886
dc.description.abstractMost multiobjective evolutionary algorithms (MOEAs) assume that analytical functions or simulation models are available while solving a multiobjective optimization problem (MOP). However, in some cases we must start with data and build approximation models known as surrogates that are later used to solve the MOP by an MOEA. These types of problems are called data-driven MOPs. This thesis is devoted to solving so-called offline data-driven MOPs that are particularly challenging as no new data is available during the optimization process. The author first presents approaches to utilize the uncertainty in the prediction of Kriging or Gaussian process (GP) surrogates as additional objectives. However, these approaches increase the complexity of the MOP being solved. Hence, the author proposes probabilistic selection approaches that can be embedded in a decomposition-based MOEA without further analytical derivations. These approaches utilize Monte Carlo sampling and kernel density estimation to calculate the probability of selection criterion of the MOEA and later select individuals based on them. Next, the author proposes an interactive optimization framework that utilizes decision maker’s preferences for uncertainties in addition to preferences for objective values. The framework was further extended to use probabilistic selection approaches for a decomposition-based MOEA and a custom reference vector adaptation technique to consider uncertainty in the solutions during the adaptation process. Building GPs with all the provided data becomes computationally expensive when the size of the data is large. Hence, the author finally proposes treed GP surrogates for multiobjective optimization (TGP-MO). They can be built with a relatively low computational cost and have a good accuracy exclusively in the regions around the optimal solutions. This thesis provides multiple novel approaches and detailed experimental studies for solving offline data-driven MOPs with decision support that will enhance real-world problem-solving capabilities. Keywords: metamodelling, surrogates, Pareto optimality, Kriging, Gaussian processes, evolutionary algorithm, decision making, uncertainty, interactive methods, preference informationen
dc.format.mimetypeapplication/pdf
dc.language.isoeng
dc.publisherJyväskylän yliopisto
dc.relation.ispartofseriesJYU dissertations
dc.relation.haspart<b>Artikkeli I:</b> Mazumdar, A., Chugh, T., Miettinen, K., & López-Ibáñez, M. (2019). On Dealing with Uncertainties from Kriging Models in Offline Data-Driven Evolutionary Multiobjective Optimization. In <i>K. Deb, E. Goodman, C. A. C. Coello, K. Klamroth, K. Miettinen, S. Mostaghim, & P. Reed (Eds.), Evolutionary Multi-Criterion Optimization : 10th International Conference, EMO 2019, East Lansing, MI, USA, March 10-13, 2019, Proceedings (pp. 463-474). Springer International Publishing. Lecture Notes in Computer Science, 11411.</i> DOI: <a href="https://doi.org/10.1007/978-3-030-12598-1_37"target="_blank"> 10.1007/978-3-030-12598-1_37</a>. JYX: <a href="https://jyx.jyu.fi/handle/123456789/63438"target="_blank"> jyx.jyu.fi/handle/123456789/63438</a>
dc.relation.haspart<b>Artikkeli II:</b> Mazumdar, Atanu; Chugh, Tinkle; Hakanen, Jussi and Miettinen, Kaisa. Prob- abilistic Selection Approaches in Decomposition-based Evolutionary Algorithms for Offline Data-Driven Multiobjective Optimization. <i>Conditionally accepted in IEEE Transactions on Evolutionary Computation.</i>
dc.relation.haspart<b>Artikkeli III:</b> Mazumdar, A., Chugh, T., Hakanen, J., & Miettinen, K. (2020). An Interactive Framework for Offline Data-Driven Multiobjective Optimization. In <i>B. Filipic, E. Minisci, & M. Vasilei (Eds.), BIOMA 2020 : 9th International Conference on Bioinspired Optimization Methods and Their Applications, Proceedings (pp. 97-109). Springer. Lecture Notes in Computer Science, 12438.</i> DOI: <a href="https://doi.org/10.1007/978-3-030-63710-1_8"target="_blank"> 10.1007/978-3-030-63710-1_8</a>. JYX: <a href=" https://jyx.jyu.fi/handle/123456789/72822"target="_blank"> jyx.jyu.fi/handle/123456789/72822</a>
dc.relation.haspart<b>Artikkeli IV:</b> Mazumdar, Atanu; López-Ibáñez, Manuel; Chugh, Tinkle; Hakanen, Jussi and Miettinen, Kaisa. TGP-MO: Treed Gaussian Processes for Solving Offline Data-Driven Multiobjective Optimization Problems. <i>Submitted to a journal.</i>
dc.rightsIn Copyright
dc.titleNovel Approaches for Offline Data-Driven Evolutionary Multiobjective Optimization
dc.typeDiss.
dc.identifier.urnURN:ISBN:978-951-39-8919-4
dc.contributor.tiedekuntaFaculty of Information Technologyen
dc.contributor.tiedekuntaInformaatioteknologian tiedekuntafi
dc.contributor.yliopistoUniversity of Jyväskyläen
dc.contributor.yliopistoJyväskylän yliopistofi
dc.relation.issn2489-9003
dc.rights.copyright© The Author & University of Jyväskylä
dc.rights.accesslevelopenAccess
dc.type.publicationdoctoralThesis
dc.format.contentfulltext
dc.rights.urlhttps://rightsstatements.org/page/InC/1.0/


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

In Copyright
Except where otherwise noted, this item's license is described as In Copyright