Deep Optimal Sensor Placement for Black Box Stochastic Simulations

Paula Cordero Encinar, Tobias Schröder, Peter Yatsyshin, Andrew B. Duncan
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:2422-2430, 2025.

Abstract

Selecting cost-effective optimal sensor configurations for subsequent inference of parameters in black-box stochastic systems faces significant computational barriers. We propose a novel and robust approach, modelling the joint distribution over input parameters and solution with a joint energy-based model, trained on simulation data. Unlike existing simulation-based inference approaches, which must be tied to a specific set of point evaluations, we learn a functional representation of parameters and solution. This is used as a resolution-independent plug-and-play surrogate for the joint distribution, which can be conditioned over any set of points, permitting an efficient approach to sensor placement. We demonstrate the validity of our framework on a variety of stochastic problems, showing that our method provides highly informative sensor locations at a lower computational cost compared to conventional approaches.

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-encinar25a, title = {Deep Optimal Sensor Placement for Black Box Stochastic Simulations}, author = {Encinar, Paula Cordero and Schr{\"o}der, Tobias and Yatsyshin, Peter and Duncan, Andrew B.}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {2422--2430}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/encinar25a/encinar25a.pdf}, url = {https://proceedings.mlr.press/v258/encinar25a.html}, abstract = {Selecting cost-effective optimal sensor configurations for subsequent inference of parameters in black-box stochastic systems faces significant computational barriers. We propose a novel and robust approach, modelling the joint distribution over input parameters and solution with a joint energy-based model, trained on simulation data. Unlike existing simulation-based inference approaches, which must be tied to a specific set of point evaluations, we learn a functional representation of parameters and solution. This is used as a resolution-independent plug-and-play surrogate for the joint distribution, which can be conditioned over any set of points, permitting an efficient approach to sensor placement. We demonstrate the validity of our framework on a variety of stochastic problems, showing that our method provides highly informative sensor locations at a lower computational cost compared to conventional approaches.} }
Endnote
%0 Conference Paper %T Deep Optimal Sensor Placement for Black Box Stochastic Simulations %A Paula Cordero Encinar %A Tobias Schröder %A Peter Yatsyshin %A Andrew B. Duncan %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-encinar25a %I PMLR %P 2422--2430 %U https://proceedings.mlr.press/v258/encinar25a.html %V 258 %X Selecting cost-effective optimal sensor configurations for subsequent inference of parameters in black-box stochastic systems faces significant computational barriers. We propose a novel and robust approach, modelling the joint distribution over input parameters and solution with a joint energy-based model, trained on simulation data. Unlike existing simulation-based inference approaches, which must be tied to a specific set of point evaluations, we learn a functional representation of parameters and solution. This is used as a resolution-independent plug-and-play surrogate for the joint distribution, which can be conditioned over any set of points, permitting an efficient approach to sensor placement. We demonstrate the validity of our framework on a variety of stochastic problems, showing that our method provides highly informative sensor locations at a lower computational cost compared to conventional approaches.
APA
Encinar, P.C., Schröder, T., Yatsyshin, P. & Duncan, A.B.. (2025). Deep Optimal Sensor Placement for Black Box Stochastic Simulations. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:2422-2430 Available from https://proceedings.mlr.press/v258/encinar25a.html.

Related Material