Simulation-based inference for rapid Bayesian parameter estimation in epidemiological models: a comparison with MCMC

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Alina Bazarova, Johann Fredrik Jadebeck, Henrik Zunker, Carolina J. Klett-Tammen, Torben Heinsohn, Wolfgang Wiechert, Katharina Noeh, Stefan Kesselheim
• Venue: arXiv: cs.AI
• Published: 2026-06-25
• Collected: 2026-06-26
• URL: http://arxiv.org/abs/2606.27286v1
• PDF: https://arxiv.org/pdf/2606.27286v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers compare Simulation-Based Inference (SBI) with Markov chain Monte Carlo (MCMC) methods for Bayesian calibration of a mechanistic epidemiological model using COVID-19 ICU occupancy data from Germany in 2020. SBI is found to be computationally more efficient, reducing runtime significantly compared to MCMC, while maintaining accuracy and posterior agreement.

What the Document Says

• The study investigates Simulation-Based Inference (SBI) as a scalable alternative for Bayesian calibration of a mechanistic SECIR epidemiological model using COVID-19 ICU data from Germany in 2020.
• SBI is compared with MCMC across multiple epidemic phases and inference windows, including a 31-day window and a 201-day reconstruction problem.
• Quantitative evaluation of posterior agreement was done using Wasserstein distances, Kullback-Leibler divergences, and posterior predictive checks.
• In the 31-day window, SBI recovered posterior distributions in strong agreement with MCMC and accurately reproduced observed ICU trajectories.
• In the 201-day setting, SBI preserved the dominant posterior structure despite increased uncertainty.
• SBI reduced computational runtime significantly compared to MCMC, taking approximately 60-70 seconds on a single GPU for the 31-day inference problems compared to about 1000 seconds for MCMC.
• For the 201-day inference problem, SBI required an average of 157 seconds, while MCMC runs took over 19,000 seconds.
• The results suggest that SBI provides a rapid and computationally efficient framework for Bayesian calibration of mechanistic epidemiological models, supporting repeated near-real-time inference and rapid outbreak analysis.

Key Concepts Mentioned

• Simulation-Based Inference (SBI)
• Markov chain Monte Carlo (MCMC)
• Bayesian calibration
• mechanistic SECIR epidemiological model
• COVID-19 ICU occupancy data
• Germany
• Wasserstein distances
• Kullback-Leibler divergences
• computational efficiency

Claims Made by the Document

• SBI provides a rapid and computationally efficient framework for Bayesian calibration of mechanistic epidemiological models.
• SBI is a scalable alternative for Bayesian calibration compared to MCMC.

Evidence Used

• SBI recovered posterior distributions in strong agreement with MCMC and accurately reproduced observed ICU trajectories.
• SBI reduced computational runtime significantly compared to MCMC, taking less time for the same inference problems.
• The average computational time for SBI was significantly lower than that of MCMC for the 201-day inference problem.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Mechanistic epidemiological models are widely used to support infectious disease forecasting and public-health decision making. Bayesian calibration of such models is commonly performed using Markov chain Monte Carlo (MCMC), which can become computationally expensive for high-dimensional nonlinear systems and repeated near-real-time analyses. Here, we investigate simulation-based inference (SBI) using neural posterior estimation as a scalable alternative for Bayesian calibration of a mechanistic SECIR epidemiological model using COVID-19 intensive care unit (ICU) occupancy data from Germany during 2020. We compared SBI and MCMC across multiple epidemic phases using both 31-day inference windows and a substantially more challenging 201-day reconstruction problem involving multiple transmission change points. Posterior agreement was evaluated quantitatively using Wasserstein distances and Kullback-Leibler divergences together with posterior predictive checks. Across the 31-day windows, SBI recovered posterior distributions in strong agreement with MCMC while accurately reproducing observed ICU trajectories. In the 201-day setting, SBI preserved the dominant posterior structure despite increased uncertainty. SBI, by combining CPU and GPU resources, substantially reduced computational runtime compared with MCMC, which was restricted to running on CPUs. Whereas MCMC required approximately 1000 seconds for the 31-day inference problems, SBI achieved comparable posterior and predictive performance in approximately 60-70 seconds on a single GPU. For the 201-day inference problem, SBI required an average of 157 seconds, while the MCMC runs took over 19,000 seconds. Our results demonstrate that SBI provides a rapid and computationally efficient framework for Bayesian calibration of mechanistic epidemiological models, supporting repeated near-real-time inference and rapid outbreak analysis.