Cost-aware simulation-based inference

Ayush Bharti, Daolang Huang, Samuel Kaski, Francois-Xavier Briol
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:28-36, 2025.

Abstract

Simulation-based inference (SBI) is rapidly becoming the preferred framework for estimating parameters of intractable models in science and engineering. A significant challenge in this context is the large computational cost of simulating data from complex models, and the fact that this cost often depends on parameter values. We therefore propose \emph{cost-aware SBI methods} which can significantly reduce the cost of existing sampling-based SBI methods, such as neural SBI and approximate Bayesian computation. This is achieved through a combination of rejection and self-normalised importance sampling, which significantly reduces the number of expensive simulations needed. Our approach is studied extensively on models from epidemiology to telecommunications engineering, where we obtain significant reductions in the overall cost of inference.

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-bharti25a, title = {Cost-aware simulation-based inference}, author = {Bharti, Ayush and Huang, Daolang and Kaski, Samuel and Briol, Francois-Xavier}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {28--36}, 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/bharti25a/bharti25a.pdf}, url = {https://proceedings.mlr.press/v258/bharti25a.html}, abstract = {Simulation-based inference (SBI) is rapidly becoming the preferred framework for estimating parameters of intractable models in science and engineering. A significant challenge in this context is the large computational cost of simulating data from complex models, and the fact that this cost often depends on parameter values. We therefore propose \emph{cost-aware SBI methods} which can significantly reduce the cost of existing sampling-based SBI methods, such as neural SBI and approximate Bayesian computation. This is achieved through a combination of rejection and self-normalised importance sampling, which significantly reduces the number of expensive simulations needed. Our approach is studied extensively on models from epidemiology to telecommunications engineering, where we obtain significant reductions in the overall cost of inference.} }
Endnote
%0 Conference Paper %T Cost-aware simulation-based inference %A Ayush Bharti %A Daolang Huang %A Samuel Kaski %A Francois-Xavier Briol %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-bharti25a %I PMLR %P 28--36 %U https://proceedings.mlr.press/v258/bharti25a.html %V 258 %X Simulation-based inference (SBI) is rapidly becoming the preferred framework for estimating parameters of intractable models in science and engineering. A significant challenge in this context is the large computational cost of simulating data from complex models, and the fact that this cost often depends on parameter values. We therefore propose \emph{cost-aware SBI methods} which can significantly reduce the cost of existing sampling-based SBI methods, such as neural SBI and approximate Bayesian computation. This is achieved through a combination of rejection and self-normalised importance sampling, which significantly reduces the number of expensive simulations needed. Our approach is studied extensively on models from epidemiology to telecommunications engineering, where we obtain significant reductions in the overall cost of inference.
APA
Bharti, A., Huang, D., Kaski, S. & Briol, F.. (2025). Cost-aware simulation-based inference. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:28-36 Available from https://proceedings.mlr.press/v258/bharti25a.html.

Related Material