FLARE MCMC: Fidelity-Based Layer-Adaptive Recursive Proposals for MCMC (2026)

by Harini Venkatesan, Christian Shelton, Ming-Feng Ho, Simeon Bird, and Mengxuan Wu

Abstract: Markov chain Monte Carlo (MCMC) requires only the ability to evaluate the likelihood, making it a common technique for inference in complex models. However, it can have a slow mixing rate, requiring the generation of many samples to obtain good estimates and an overall high computational cost. FLARE MCMC is a multi-fidelity layered MCMC method that exploits lower-fidelity approximations of the true likelihood calculation to improve mixing and leads to overall faster performance. Such lower-fidelity likelihoods are commonly available in scientific and engineering applications where the model involves a simulation whose resolution or accuracy can be tuned. Our technique uses recursive, layered chains with simple layer tuning; it does not require the likelihood to take any form or have any particular internal mathematical structure. We demonstrate experimentally that FLARE MCMC achieves larger effective sample sizes for the same computational time across different scientific domains including hydrology and cosmology.

Download Information

Harini Venkatesan, Christian Shelton, Ming-Feng Ho, Simeon Bird, and Mengxuan Wu (2026). "FLARE MCMC: Fidelity-Based Layer-Adaptive Recursive Proposals for MCMC." SIAM/ASA Journal on Uncertainty Quantification, 14(2), 661-690.

Bibtex citation

@article{Venetal26,
     author = {Harini Venkatesan and Christian Shelton and Ming-Feng Ho and Simeon Bird and Mengxuan Wu},
     title = "{FLARE MCMC}: Fidelity-Based Layer-Adaptive Recursive Proposals for {MCMC}",
     journal = "{SIAM/ASA} Journal on Uncertainty Quantification",
     journalabbr = "JUQ",
     volume = 14,
     number = 2,
     pages = "661--690",
     year = 2026,
     doi = "10.1137/25M1795194",
}

---

full list