Home > Research > Publications & Outputs > Sequential Neural Score Estimation

Research

Associated organisational unit

Electronic data

  • 9103_Sequential_Neural_Score_E

    Accepted author manuscript, 2.73 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

View graph of relations

Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models

Research output: Contribution to Journal/Magazine › Conference article › peer-review

Published

Standard

Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models. / Sharrock, Louis; Simons, Jack; Liu, Song et al.
In: Proceedings of Machine Learning Research, Vol. 235, 29.07.2024, p. 44565-44602.

Research output: Contribution to Journal/Magazine › Conference article › peer-review

Harvard

Sharrock, L, Simons, J, Liu, S & Beaumont, M 2024, 'Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models', Proceedings of Machine Learning Research, vol. 235, pp. 44565-44602. <https://openreview.net/pdf?id=8viuf9PdzU>

APA

Sharrock, L., Simons, J., Liu, S., & Beaumont, M. (2024). Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models. Proceedings of Machine Learning Research, 235, 44565-44602. https://openreview.net/pdf?id=8viuf9PdzU

Vancouver

Sharrock L, Simons J, Liu S, Beaumont M. Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models. Proceedings of Machine Learning Research. 2024 Jul 29;235:44565-44602.

Author

Sharrock, Louis ; Simons, Jack ; Liu, Song et al. / Sequential Neural Score Estimation : Likelihood-Free Inference with Conditional Score Based Diffusion Models. In: Proceedings of Machine Learning Research. 2024 ; Vol. 235. pp. 44565-44602.

Bibtex

@article{02279c3f852643cab5b50271c1d6f10c,
title = "Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models",
abstract = "We introduce Sequential Neural Posterior Score Estimation (SNPSE), a score-based method for Bayesian inference in simulator-based models. Our method, inspired by the remarkable success of score-based methods in generative modelling, leverages conditional score-based diffusion models to generate samples from the posterior distribution of interest. The model is trained using an objective function which directly estimates the score of the posterior. We embed the model into a sequential training procedure, which guides simulations using the current approximation of the posterior at the observation of interest, thereby reducing the simulation cost. We also introduce several alternative sequential approaches, and discuss their relative merits. We then validate our method, as well as its amortised, non-sequential, variant on several numerical examples, demonstrating comparable or superior performance to existing state-of-the-art methods such as Sequential Neural Posterior Estimation (SNPE).",
author = "Louis Sharrock and Jack Simons and Song Liu and Mark Beaumont",
note = "In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), Vienna, Austria. ",
year = "2024",
month = jul,
day = "29",
language = "English",
volume = "235",
pages = "44565--44602",
journal = "Proceedings of Machine Learning Research",
issn = "1938-7228",
publisher = "ML Research Press",

}

RIS

TY - JOUR

T1 - Sequential Neural Score Estimation

T2 - Likelihood-Free Inference with Conditional Score Based Diffusion Models

AU - Sharrock, Louis

AU - Simons, Jack

AU - Liu, Song

AU - Beaumont, Mark

N1 - In: Proceedings of the 41st International Conference on Machine Learning (ICML 2024), Vienna, Austria.

PY - 2024/7/29

Y1 - 2024/7/29

N2 - We introduce Sequential Neural Posterior Score Estimation (SNPSE), a score-based method for Bayesian inference in simulator-based models. Our method, inspired by the remarkable success of score-based methods in generative modelling, leverages conditional score-based diffusion models to generate samples from the posterior distribution of interest. The model is trained using an objective function which directly estimates the score of the posterior. We embed the model into a sequential training procedure, which guides simulations using the current approximation of the posterior at the observation of interest, thereby reducing the simulation cost. We also introduce several alternative sequential approaches, and discuss their relative merits. We then validate our method, as well as its amortised, non-sequential, variant on several numerical examples, demonstrating comparable or superior performance to existing state-of-the-art methods such as Sequential Neural Posterior Estimation (SNPE).

AB - We introduce Sequential Neural Posterior Score Estimation (SNPSE), a score-based method for Bayesian inference in simulator-based models. Our method, inspired by the remarkable success of score-based methods in generative modelling, leverages conditional score-based diffusion models to generate samples from the posterior distribution of interest. The model is trained using an objective function which directly estimates the score of the posterior. We embed the model into a sequential training procedure, which guides simulations using the current approximation of the posterior at the observation of interest, thereby reducing the simulation cost. We also introduce several alternative sequential approaches, and discuss their relative merits. We then validate our method, as well as its amortised, non-sequential, variant on several numerical examples, demonstrating comparable or superior performance to existing state-of-the-art methods such as Sequential Neural Posterior Estimation (SNPE).

M3 - Conference article

VL - 235

SP - 44565

EP - 44602

JO - Proceedings of Machine Learning Research

JF - Proceedings of Machine Learning Research

SN - 1938-7228

ER -