Current studentsExisting staffContact us
 

Pareto set representation learning with application to multi-criteria order optimization

Simple item page
dc.contributor.authorTan, C.S.
dc.contributor.authorGupta, A.
dc.contributor.authorOng, Y.S.
dc.contributor.authorLam, S.K.
dc.contributor.authorPratama, M.
dc.contributor.authorTan, P.S.
dc.contributor.conference2024 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM) (15 Dec 2024 - 18 Dec 2024 : Bangkok, Thailand)
dc.date.issued2024
dc.description.abstractMulti-objective optimization seeks to arrive at a diverse set of Pareto-optimal solutions facilitating a posteriori decision-making. However, this becomes challenging for high-dimensional problems with limited compute, imposing a compromise between convergence and diversity of the final solutions. To address this curse of dimensionality, we introduce the concept of Pareto set representation learning, reducing the problem to its smallest possible dimensions while accurately capturing the Pareto-optima. A denoising autoencoder is invoked to discover a compressed latent representation of a sparsely populated Pareto set by leveraging its unique bottleneck architecture. This representation then serves as a means to create compact inverse models, mapping points from the Pareto front in objective space to the (dimensionally reduced) Pareto set in decision space. The method is empirically tested on benchmark problems and an industrial multi-site order planning problem showcasing its effectiveness in reducing the dimensionality of the Pareto set (99.6%) while achieving significant gains (>200%) in Pareto approximation capacity. With such compact yet accurate inverse models, decision makers can readily generate high-dimensional solutions corresponding to any preferred, unexplored subregions of the objective space.
dc.identifier.citationIEEE International Conference on Industrial Engineering and Engineering Management, 2024, pp.947-951
dc.identifier.doi10.1109/IEEM62345.2024.10857217
dc.identifier.isbn9798350386103
dc.identifier.issn2157-3611
dc.identifier.issn2157-362X
dc.identifier.urihttps://hdl.handle.net/11541.2/41962
dc.language.isoen
dc.publisherIEEE
dc.publisher.placeUS
dc.relation.fundingSERB RJF/2022/000115
dc.relation.fundingNTU
dc.rightsCopyright 2024 IEEE
dc.source.urihttps://doi.org/10.1109/IEEM62345.2024.10857217
dc.subjectdenoising autoencoder
dc.subjectmulti-objective optimization
dc.subjectmulti-site order planning
dc.subjectpareto set representation learning
dc.titlePareto set representation learning with application to multi-criteria order optimization
dc.typeConference paper
pubs.publication-statusPublished
ror.mmsid9916946117601831

Files

Collections