Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design

Marcel Hedman, Desi R. Ivanova, Cong Guan, Tom Rainforth
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:22904-22923, 2025.

Abstract

We develop a semi-amortized, policy-based, approach to Bayesian experimental design (BED) called Stepwise Deep Adaptive Design (Step-DAD). Like existing, fully amortized, policy-based BED approaches, Step-DAD trains a design policy upfront before the experiment. However, rather than keeping this policy fixed, Step-DAD periodically updates it as data is gathered, refining it to the particular experimental instance. This test-time adaptation improves both the flexibility and the robustness of the design strategy compared with existing approaches. Empirically, Step-DAD consistently demonstrates superior decision-making and robustness compared with current state-of-the-art BED methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-hedman25a, title = {Step-{DAD}: Semi-Amortized Policy-Based {B}ayesian Experimental Design}, author = {Hedman, Marcel and Ivanova, Desi R. and Guan, Cong and Rainforth, Tom}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {22904--22923}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/hedman25a/hedman25a.pdf}, url = {https://proceedings.mlr.press/v267/hedman25a.html}, abstract = {We develop a semi-amortized, policy-based, approach to Bayesian experimental design (BED) called Stepwise Deep Adaptive Design (Step-DAD). Like existing, fully amortized, policy-based BED approaches, Step-DAD trains a design policy upfront before the experiment. However, rather than keeping this policy fixed, Step-DAD periodically updates it as data is gathered, refining it to the particular experimental instance. This test-time adaptation improves both the flexibility and the robustness of the design strategy compared with existing approaches. Empirically, Step-DAD consistently demonstrates superior decision-making and robustness compared with current state-of-the-art BED methods.} }
Endnote
%0 Conference Paper %T Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design %A Marcel Hedman %A Desi R. Ivanova %A Cong Guan %A Tom Rainforth %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-hedman25a %I PMLR %P 22904--22923 %U https://proceedings.mlr.press/v267/hedman25a.html %V 267 %X We develop a semi-amortized, policy-based, approach to Bayesian experimental design (BED) called Stepwise Deep Adaptive Design (Step-DAD). Like existing, fully amortized, policy-based BED approaches, Step-DAD trains a design policy upfront before the experiment. However, rather than keeping this policy fixed, Step-DAD periodically updates it as data is gathered, refining it to the particular experimental instance. This test-time adaptation improves both the flexibility and the robustness of the design strategy compared with existing approaches. Empirically, Step-DAD consistently demonstrates superior decision-making and robustness compared with current state-of-the-art BED methods.
APA
Hedman, M., Ivanova, D.R., Guan, C. & Rainforth, T.. (2025). Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:22904-22923 Available from https://proceedings.mlr.press/v267/hedman25a.html.

Related Material