Beyond the Pareto Efficient Frontier: Constraint Active Search for Multiobjective Experimental Design

Gustavo Malkomes, Bolong Cheng, Eric H Lee, Mike Mccourt
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:7423-7434, 2021.

Abstract

Many problems in engineering design and simulation require balancing competing objectives under the presence of uncertainty. Sample-efficient multiobjective optimization methods focus on the objective function values in metric space and ignore the sampling behavior of the design configurations in parameter space. Consequently, they may provide little actionable insight on how to choose designs in the presence of metric uncertainty or limited precision when implementing a chosen design. We propose a new formulation that accounts for the importance of the parameter space and is thus more suitable for multiobjective design problems; instead of searching for the Pareto-efficient frontier, we solicit the desired minimum performance thresholds on all objectives to define regions of satisfaction. We introduce an active search algorithm called Expected Coverage Improvement (ECI) to efficiently discover the region of satisfaction and simultaneously sample diverse acceptable configurations. We demonstrate our algorithm on several design and simulation domains: mechanical design, additive manufacturing, medical monitoring, and plasma physics.

Cite this Paper


BibTeX
@InProceedings{pmlr-v139-malkomes21a, title = {Beyond the Pareto Efficient Frontier: Constraint Active Search for Multiobjective Experimental Design}, author = {Malkomes, Gustavo and Cheng, Bolong and Lee, Eric H and Mccourt, Mike}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {7423--7434}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/malkomes21a/malkomes21a.pdf}, url = {https://proceedings.mlr.press/v139/malkomes21a.html}, abstract = {Many problems in engineering design and simulation require balancing competing objectives under the presence of uncertainty. Sample-efficient multiobjective optimization methods focus on the objective function values in metric space and ignore the sampling behavior of the design configurations in parameter space. Consequently, they may provide little actionable insight on how to choose designs in the presence of metric uncertainty or limited precision when implementing a chosen design. We propose a new formulation that accounts for the importance of the parameter space and is thus more suitable for multiobjective design problems; instead of searching for the Pareto-efficient frontier, we solicit the desired minimum performance thresholds on all objectives to define regions of satisfaction. We introduce an active search algorithm called Expected Coverage Improvement (ECI) to efficiently discover the region of satisfaction and simultaneously sample diverse acceptable configurations. We demonstrate our algorithm on several design and simulation domains: mechanical design, additive manufacturing, medical monitoring, and plasma physics.} }
Endnote
%0 Conference Paper %T Beyond the Pareto Efficient Frontier: Constraint Active Search for Multiobjective Experimental Design %A Gustavo Malkomes %A Bolong Cheng %A Eric H Lee %A Mike Mccourt %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-malkomes21a %I PMLR %P 7423--7434 %U https://proceedings.mlr.press/v139/malkomes21a.html %V 139 %X Many problems in engineering design and simulation require balancing competing objectives under the presence of uncertainty. Sample-efficient multiobjective optimization methods focus on the objective function values in metric space and ignore the sampling behavior of the design configurations in parameter space. Consequently, they may provide little actionable insight on how to choose designs in the presence of metric uncertainty or limited precision when implementing a chosen design. We propose a new formulation that accounts for the importance of the parameter space and is thus more suitable for multiobjective design problems; instead of searching for the Pareto-efficient frontier, we solicit the desired minimum performance thresholds on all objectives to define regions of satisfaction. We introduce an active search algorithm called Expected Coverage Improvement (ECI) to efficiently discover the region of satisfaction and simultaneously sample diverse acceptable configurations. We demonstrate our algorithm on several design and simulation domains: mechanical design, additive manufacturing, medical monitoring, and plasma physics.
APA
Malkomes, G., Cheng, B., Lee, E.H. & Mccourt, M.. (2021). Beyond the Pareto Efficient Frontier: Constraint Active Search for Multiobjective Experimental Design. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:7423-7434 Available from https://proceedings.mlr.press/v139/malkomes21a.html.

Related Material