PASOA- PArticle baSed Bayesian Optimal Adaptive design

Jacopo Iollo, Christophe Heinkelé, Pierre Alliez, Florence Forbes
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:21020-21046, 2024.

Abstract

We propose a new procedure named PASOA, for Bayesian experimental design, that performs sequential design optimization by simultaneously providing accurate estimates of successive posterior distributions for parameter inference. The sequential design process is carried out via a contrastive estimation principle, using stochastic optimization and Sequential Monte Carlo (SMC) samplers to maximise the Expected Information Gain (EIG). As larger information gains are obtained for larger distances between successive posterior distributions, this EIG objective may worsen classical SMC performance. To handle this issue, tempering is proposed to have both a large information gain and an accurate SMC sampling, that we show is crucial for performance. This novel combination of stochastic optimization and tempered SMC allows to jointly handle design optimization and parameter inference. We provide a proof that the obtained optimal design estimators benefit from some consistency property. Numerical experiments confirm the potential of the approach, which outperforms other recent existing procedures.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-iollo24a, title = {{PASOA}- {PA}rticle ba{S}ed {B}ayesian Optimal Adaptive design}, author = {Iollo, Jacopo and Heinkel\'{e}, Christophe and Alliez, Pierre and Forbes, Florence}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {21020--21046}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/iollo24a/iollo24a.pdf}, url = {https://proceedings.mlr.press/v235/iollo24a.html}, abstract = {We propose a new procedure named PASOA, for Bayesian experimental design, that performs sequential design optimization by simultaneously providing accurate estimates of successive posterior distributions for parameter inference. The sequential design process is carried out via a contrastive estimation principle, using stochastic optimization and Sequential Monte Carlo (SMC) samplers to maximise the Expected Information Gain (EIG). As larger information gains are obtained for larger distances between successive posterior distributions, this EIG objective may worsen classical SMC performance. To handle this issue, tempering is proposed to have both a large information gain and an accurate SMC sampling, that we show is crucial for performance. This novel combination of stochastic optimization and tempered SMC allows to jointly handle design optimization and parameter inference. We provide a proof that the obtained optimal design estimators benefit from some consistency property. Numerical experiments confirm the potential of the approach, which outperforms other recent existing procedures.} }
Endnote
%0 Conference Paper %T PASOA- PArticle baSed Bayesian Optimal Adaptive design %A Jacopo Iollo %A Christophe Heinkelé %A Pierre Alliez %A Florence Forbes %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-iollo24a %I PMLR %P 21020--21046 %U https://proceedings.mlr.press/v235/iollo24a.html %V 235 %X We propose a new procedure named PASOA, for Bayesian experimental design, that performs sequential design optimization by simultaneously providing accurate estimates of successive posterior distributions for parameter inference. The sequential design process is carried out via a contrastive estimation principle, using stochastic optimization and Sequential Monte Carlo (SMC) samplers to maximise the Expected Information Gain (EIG). As larger information gains are obtained for larger distances between successive posterior distributions, this EIG objective may worsen classical SMC performance. To handle this issue, tempering is proposed to have both a large information gain and an accurate SMC sampling, that we show is crucial for performance. This novel combination of stochastic optimization and tempered SMC allows to jointly handle design optimization and parameter inference. We provide a proof that the obtained optimal design estimators benefit from some consistency property. Numerical experiments confirm the potential of the approach, which outperforms other recent existing procedures.
APA
Iollo, J., Heinkelé, C., Alliez, P. & Forbes, F.. (2024). PASOA- PArticle baSed Bayesian Optimal Adaptive design. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:21020-21046 Available from https://proceedings.mlr.press/v235/iollo24a.html.

Related Material