Constrained Bayesian Experimental Design via Online Planning

Yujia Guo, Daolang Huang, Xinyu Zhang, Sammie Katt, Samuel Kaski, Ayush Bharti
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:37961-37984, 2026.

Abstract

Bayesian experimental design (BED) is a principled framework for data-efficient design of sequential experiments. However, existing BED methods are unable to adapt to dynamic constraints inherent in real-world tasks due to budget limitations, varying costs, or physical constraints that restrict how designs evolve over time. In this paper, we introduce a novel approach to BED that enables constrained optimization of experimental designs by combining offline pre-training of an amortized policy and a posterior network with online multi-step lookahead planning using scenario trees. We empirically demonstrate that our method yields substantially more informative design sequences than existing methods across a range of constrained BED tasks, while incurring only a modest additional computational overhead.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-guo26g, title = {Constrained {B}ayesian Experimental Design via Online Planning}, author = {Guo, Yujia and Huang, Daolang and Zhang, Xinyu and Katt, Sammie and Kaski, Samuel and Bharti, Ayush}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {37961--37984}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/guo26g/guo26g.pdf}, url = {https://proceedings.mlr.press/v306/guo26g.html}, abstract = {Bayesian experimental design (BED) is a principled framework for data-efficient design of sequential experiments. However, existing BED methods are unable to adapt to dynamic constraints inherent in real-world tasks due to budget limitations, varying costs, or physical constraints that restrict how designs evolve over time. In this paper, we introduce a novel approach to BED that enables constrained optimization of experimental designs by combining offline pre-training of an amortized policy and a posterior network with online multi-step lookahead planning using scenario trees. We empirically demonstrate that our method yields substantially more informative design sequences than existing methods across a range of constrained BED tasks, while incurring only a modest additional computational overhead.} }
Endnote
%0 Conference Paper %T Constrained Bayesian Experimental Design via Online Planning %A Yujia Guo %A Daolang Huang %A Xinyu Zhang %A Sammie Katt %A Samuel Kaski %A Ayush Bharti %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-guo26g %I PMLR %P 37961--37984 %U https://proceedings.mlr.press/v306/guo26g.html %V 306 %X Bayesian experimental design (BED) is a principled framework for data-efficient design of sequential experiments. However, existing BED methods are unable to adapt to dynamic constraints inherent in real-world tasks due to budget limitations, varying costs, or physical constraints that restrict how designs evolve over time. In this paper, we introduce a novel approach to BED that enables constrained optimization of experimental designs by combining offline pre-training of an amortized policy and a posterior network with online multi-step lookahead planning using scenario trees. We empirically demonstrate that our method yields substantially more informative design sequences than existing methods across a range of constrained BED tasks, while incurring only a modest additional computational overhead.
APA
Guo, Y., Huang, D., Zhang, X., Katt, S., Kaski, S. & Bharti, A.. (2026). Constrained Bayesian Experimental Design via Online Planning. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:37961-37984 Available from https://proceedings.mlr.press/v306/guo26g.html.

Related Material