Pith. sign in

REVIEW 5 major objections 4 minor 43 references

Modern Bayesian Sampling Methods for Cosmological Inference: A Comparative Study

T0 review · 5 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read No single Bayesian sampler is universally superior; Hamiltonian Monte Carlo offers the best speed–accuracy trade-off for higher-dimensional cosmological inference, while nested samplers justify their higher cost by computing the evidence…

desk verdict A useful but uneven sampler benchmark on a realistic BAO likelihood; the qualitative 'no universal winner' conclusion likely holds, but the headline HMC runtime advantage is not controlled enough to trust as a general statement. read the letter →

arxiv 2501.06022 v2 pith:MSMGMWEH submitted 2025-01-10 astro-ph.CO

Denitsa Staicova This is my paper · ORCID
classification astro-ph.CO
keywords cosmologyMarkovchainMonteCarloHamiltoniannestedsamplingslicebaryonacousticoscillationsBayesianevidencesamplerbenchmarking
topics Dark Energy
open problems Dark Energy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks which Markov chain Monte Carlo sampler cosmologists should trust when estimating parameters from data such as baryon acoustic oscillations. It runs six samplers—a traditional random-walk MCMC, Hamiltonian Monte Carlo, slice sampling, an ensemble sampler, and two nested samplers—on identical test distributions and on a cosmological likelihood with one to four free parameters. Its central finding is that no single method is universally better: all recover well-constrained parameters reliably, and the ranking depends on dimensionality, curved degeneracies, and whether the Bayesian evidence is needed. When gradients are available, Hamiltonian Monte Carlo gives the best balance of runtime and accuracy in higher dimensions, while nested samplers remain accurate at higher computational cost and directly supply the evidence.

What carries the argument

The load-bearing mechanism is the unified benchmark framework that standardizes metrics—runtime, memory usage, effective sample size per second, initialization sensitivity, and mean-based plus distribution-based accuracy—across all samplers on the same targets. Within that framework, the deciding quantities are the gradient-driven leapfrog dynamics of HMC, which preserve volume in phase space, and the live-point evidence integral of nested sampling, which converts the multi-dimensional evidence integral into a one-dimensional integration over prior mass. The test suite uses a correlated Gaussian, a Rosenbrock banana distribution, and a Gaussian mixture; the cosmological case uses a marginalized BAO likelihood with $\Omega_m$, $\Omega_K$, $w_0$, and $w_a$ as free parameters.

What would settle it

Rerun the same BAO benchmark with a matched accuracy target or a matched computational budget (for instance, requiring every sampler to reach the same effective sample size or the same evidence precision) and check whether HMC still has the best runtime while nested samplers remain the most expensive; if the rankings change under equalized settings, the reported ordering is an artifact of configuration.

Watch

Extended reading notes

Core claim

The paper claims that sampler choice in cosmological inference should be driven by problem geometry and by what the analysis needs, because the tested samplers are comparable in accuracy yet differ markedly in cost. In the test problems, all samplers handle Gaussian targets, but curved degeneracies and multimodality separate them: traditional MCMC and slice sampling lose efficiency as dimension grows, HMC holds up on curved targets, and nested samplers stay reliable on multimodal ones. On the baryon-acoustic-oscillation likelihood, HMC has the best runtime and accuracy, especially for the 3- and 4-parameter dark-energy models; nested samplers cost more but remain accurate and compute the evidence; traditional MCMC and the ensemble sampler are adequate at low dimension but widen uncertainties in higher dimension. The paper concludes that no single method is significantly better across all settings, and that choosing a sampler means weighing dimensionality, likelihood features, and computational constraints.

Load-bearing premise

The comparison assumes that the fixed per-sampler settings—such as the number of live points, walkers, warmup steps, and burn-in fractions—place the methods on an equal footing, rather than merely reflecting arbitrary configuration choices.

Editorial extensions

If this is right

  • For well-constrained parameters like $\Omega_m$, all samplers give consistent constraints, so the simplest methods are safe in low-dimensional problems.
  • Higher-dimensional dark-energy models should default to HMC when the likelihood is differentiable and the evidence is not required, since it delivers accuracy at a fraction of the runtime of nested samplers.
  • When Bayesian model comparison is the goal, nested samplers are the practical route because they output the evidence $Z$ directly, whereas MCMC evidence estimates are approximate.
  • Effective sample size per second alone can mislead: one sampler shows high ESS but larger parameter uncertainties in higher dimensions, so raw sampling efficiency does not guarantee optimal posterior exploration.
  • Relying on a single sampler family risks geometry-induced bias; the observed differences in curved and multimodal targets imply that multiple sampling approaches are worth running for non-standard parameter spaces.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural testable extension would fix a common effective-sample-size target or a common wall-clock budget across all samplers; the paper's rankings, especially HMC's runtime advantage, may depend on the chosen per-sampler settings such as 1000 versus 100 live points.
  • The paper's emphasis on geometry suggests that future comparisons should include non-differentiable or expensive likelihoods where gradients are unavailable, a condition that would erode HMC's main advantage.
  • If the no-single-method conclusion generalizes, cosmology pipelines should choose samplers by the downstream need—evidence versus posterior—and treat the two nested samplers and HMC as complementary tools rather than substitutes.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper benchmarks six samplers—traditional Metropolis-Hastings MCMC via PyMC, HMC via NumPyro, a custom slice-sampling implementation, emcee, dynesty nested sampling, and PolyChord—on three synthetic distributions (correlated Gaussian, Rosenbrock, Gaussian mixture) and on a marginalized BAO likelihood with 1–4 cosmological parameters. Performance is assessed via runtime, memory usage, effective sample size per second, initial-seed sensitivity, and two normalized accuracy metrics. The paper concludes that HMC provides the best balance of runtime and accuracy, especially in higher dimensions, and that no single sampler is significantly better overall, with nested samplers recommended when evidence computation is required.

Significance. A careful, reproducible comparison of widely used samplers on realistic cosmological likelihoods would be a valuable practical reference for the community. The paper's strengths are the use of standard test problems, a DESI-based BAO likelihood that is not trivial, and the intention to release the benchmark code. However, the central comparative claims are not currently supported by the evidence: the samplers are run under substantially different sample counts and live-point settings, accuracy normalization is arbitrary, and runtime and ESS metrics are reported without error bars or control for parallelization. If these issues are addressed, the paper could become a useful guide for sampler selection; in its present form, the conclusions are not robust.

major comments (5)
  1. [Appendix C, Section 5.2] The sampler configurations in Appendix C impose very different computational efforts: dynesty uses nlive=1000, PolyChord nlive=100, emcee uses 1000 steps, traditional MCMC uses 5000 draws, HMC uses 500 warmup steps with an unspecified number of post-warmup draws, and slice sampling uses width=0.1. No common accuracy target, effective-sample-size target, or likelihood-evaluation budget is enforced, so the runtime and ESS-per-second rankings in Figures 2 and 5 conflate algorithmic performance with configuration choices. This directly affects the central claim in Section 5.2 that HMC provides the best balance of runtime and accuracy; the benchmark should be rerun under a common budget, for example a fixed number of likelihood evaluations or a target ESS, or the choice of settings should be explicitly justified as a fair basis for comparison.
  2. [Section 3, Figure 5] Runtime, memory, and ESS per second are reported as single point values without error bars or repeated-run statistics. The 'Init sensitivity' metric tracks seed variation of parameter accuracy only, not variation of runtime or ESS. Given that the paper's main conclusions are about runtime and efficiency, the lack of variance estimates makes it impossible to judge whether differences such as the factor of ten between HMC and PolyChord in the 3D cosmological case are significant or within run-to-run noise.
  3. [Appendix A, Eqs. (A1)–(A4)] The normalized accuracy scores depend on arbitrarily fixed maximum error thresholds (0.5 for Gaussian, 2.0 for Rosenbrock, 5.0 for the mixture). Because these thresholds are not derived from any property of the target distributions, the absolute accuracy values in Figure 3 are meaningful only relative to the chosen normalization, and the rankings could change if different thresholds were adopted. The paper should either justify the thresholds or report the unnormalized errors alongside the normalized versions.
  4. [Section 4.2, Figure 3] The text states that 'the traditional method was convergent only for the Gaussian distribution,' yet Figure 3 reports traditional MCMC accuracy values for the Rosenbrock and mixture problems. If the chains did not converge, those accuracy values should be excluded or explicitly flagged as non-converged; as written, the figure contradicts the text and the comparison is unclear.
  5. [Section 5.4, Section 3] The paper acknowledges in Section 5.4 that memory usage for parallel nested samplers is difficult to measure and may be underestimated, and Section 3 notes that dynesty, PolyChord, and emcee support parallelization while the HMC NumPyro implementation appears serial. No core counts or parallel settings are reported, so the runtime and memory comparisons mix serial and parallel execution. The benchmark should specify and control the number of cores or processes used for each sampler, or at least quantify the effect of parallelism on the reported metrics.
minor comments (4)
  1. [Section 2.4 and Appendix C, Eq. (A6)] The symbol Z is used both for the Bayesian evidence and for the normalization term in the marginalized BAO likelihood (Eq. A6); using a different symbol for one of these would avoid ambiguity.
  2. [Section 3] The description '5000 draws (ndraws ×5)' is ambiguous; it should be clarified whether 5000 is the total number of draws across all chains or the number per chain.
  3. [Figure 5] The y-axis label 'Memory (MB)' includes values up to 70000; using a log scale or expressing the units in GB would improve readability.
  4. [Section 2.5] There is a typo in 'developped' that should be corrected to 'developed'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an empirical sampler benchmark whose conclusions rest on external likelihoods and measured runtimes, not on any quantity fitted and then repackaged as a prediction.

full rationale

The paper's central claims, that HMC offers a good runtime-accuracy balance and that no single sampler is universally superior, are empirical observations from running six samplers on test distributions and a BAO likelihood. No parameter is fitted to a subset of data and then presented as a prediction; no claimed first-principles result is defined in terms of the conclusion. The only author self-citations are references [26] and [27], which supply the marginalized BAO likelihood; that likelihood is a published, externally specified function based on DESI data and is reused here as a benchmark input rather than derived from the present comparison. The manuscript's own caveats in Section 5.4 about memory measurement, parallelization, and non-uniform interfaces are implementation limitations, not admissions of circularity. A reviewer could question whether the fixed per-sampler settings in Appendix C make runtime rankings configuration-dependent, but that would be a fairness or robustness criticism of an empirical benchmark, not the pattern of a prediction reducing by construction to its inputs. Accordingly, no circular step is exhibited.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central benchmark claims rest on standard sampling theory, the author's previously published marginalized BAO likelihood, and several arbitrary normalization and hyperparameter choices that are not independently justified. No new physical entities are introduced.

free parameters (2)
  • Accuracy normalization thresholds = 0.5 (Gaussian), 2.0 (Rosenbrock), 5.0 (mixture)
    Chosen by hand in Appendix A to normalize accuracy scores across problems; different thresholds would change the reported relative accuracy of samplers.
  • Sampler hyperparameter sets = dynesty nlive=1000; PolyChord nlive=100; emcee nwalkers=max(20*ndim,40); traditional MCMC 5000 draws, 50% burn-in; HMC…
    Ad hoc choices not tuned to a common accuracy target; they directly affect runtime, ESS, and accuracy comparisons, so they are effectively free parameters of the benchmark.
assumptions (4)
  • standard math Standard convergence properties of MCMC and nested sampling algorithms.
    The paper assumes that with sufficient samples and the stated stopping criteria (R-hat<1.1, ESS>100, dlogZ<0.1), chains represent the target posterior (Convergence Criteria in Appendix C).
  • domain assumption The marginalized BAO likelihood from Staicova and Benisty (2022) is a valid likelihood for cosmological parameter inference.
    Used without derivation in Section 5.1; the full likelihood details are only referenced to the author's prior papers [26,27].
  • domain assumption DESI BAO data as used in the marginalized likelihood are correct and appropriate.
    The paper takes the DESI 2024 BAO measurements [25] as input without independent validation.
  • ad hoc to paper Uniform priors on (-5,5) for test problems and the listed cosmological priors are appropriate for the benchmark.
    Priors are chosen for the benchmark; different priors could change accuracy comparisons.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Modern Bayesian Sampling Methods for Cosmological Inference: A Comparative Study." pith.science (2026). https://pith.science/paper/MSMGMWEH

@misc{pith2026250106022,
  author       = {Pith},
  title        = {Pith review of: Modern Bayesian Sampling Methods for Cosmological Inference: A Comparative Study},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MSMGMWEH}},
  note         = {Machine review of arXiv:2501.06022}
}
read the original abstract

We present a comprehensive comparison of different Markov Chain Monte Carlo (MCMC) sampling methods, evaluating their performance on both standard test problems and cosmological parameter estimation. Our analysis includes traditional Metropolis-Hastings MCMC, Hamiltonian Monte Carlo (HMC), slice sampling, nested sampling as implemented in dynesty, and PolyChord. We examine samplers through multiple metrics including runtime, memory usage, effective sample size, and parameter accuracy, testing their scaling with dimension and response to different probability distributions. While all samplers perform well with simple Gaussian distributions, we find that HMC and nested sampling show advantages for more complex distributions typical of cosmological problems. Traditional MCMC and slice sampling become less efficient in higher dimensions, while nested methods maintain accuracy but at higher computational cost. In cosmological applications using BAO data, we observe similar patterns, with particular challenges arising from parameter degeneracies and poorly constrained parameters.

Figures

Figures reproduced from arXiv: 2501.06022 by the authors.

Figure 1
Figure 1. The surface plots corresponding to the three test problems. The global maximum that the sampler needs to find is marked in the case of the Rosenbrock distribution; the two others correspond to single and double Gaussians, respectively. Correlated Gaussian: The simplest test case involves a multivariate Gaussian distribution: logL(x) = − 1 2 d ∑ i=1 x 2 i , (7) where d is the dimension of the parameter space and xi r… view at source ↗
Figure 2
Figure 2. The summary of the metrics we track for the different samplers. We show here the runtime, the memory usage, the ESS per sec, and the Init sensitivity [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Comparison between the mean accuracy and the distribution accuracy for different samplers. The normalization is described in the Appendix A. The Rosenbrock function reveals significant differences between samplers. Its curved ”banana” shape poses challenges for traditional MCMC and slice methods, evidenced by their declining ESS per second with increasing dimensions. While most samplers main￾tain reasonable mean acc… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: A slice of the density of the likelihood of the wwaCDM model when two parameters vary and the other is set to its fiducial value Ωm = 0.3, w0 = −1, wa = 0. Note that this is not a contour plot of the posterior but a direct evaluation of the likelihood function [PITH_F…
Figure 5
Figure 5. Figure 5: The summary of the benchmark on cosmological models using the different samplers. Memory requirements scale distinctly for each sampler implementation. Traditional MCMC shows significant memory use, while HMC maintains relatively stable memory usage through its computa…
Figure 6
Figure 6. Figure 6: The left panel shows the well-constrained Ωm for all models. The right panel displays deviations for the additional parameters: spatial curvature (Ωk ) in the 2D model and the 4D model, equation of state w in the 3D model, and both w and wa in the 4D model. The spatial…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 23 canonical work pages

  1. [8]

    A comparison of Bayesian sampling algorithms for high-dimensional particle physics and cosmology applications

    Albert, J.; Balazs, C.; Fowlie, A.; Handley, W.; Hunt-Smith, N.; de Austri, R.R.; White, M. A comparison of Bayesian sampling algorithms for high-dimensional particle physics and cosmology applications. arXiv 2024, arXiv:2409.18464

    arXiv 2024

  2. [1]

    Cosmological parameters from CMB and other data: A Monte Carlo approach

    Lewis, A.; Bridle, S. Cosmological parameters from CMB and other data: A Monte Carlo approach. Phys. Rev. D 2002, 66, 103511

    work page 2002

  3. [2]

    Bayes in the sky: Bayesian inference and model selection in cosmology

    Trotta, R. Bayes in the sky: Bayesian inference and model selection in cosmology. Contemp. Phys. 2008, 49, 71–104

    work page 2008

  4. [3]

    Aghanim, N.; Akrami, Y.; Ashdown, M. et al. [Planck Collaboration] Planck 2018 results: VI. Cosmological parameters. Astron. Astrophys. 2020, 641, A6

    work page 2018

  5. [4]

    Equation of state calculations by fast computing machines

    Metropolis, N.; Rosenbluth, A.W.; Rosenbluth, M.N.; Teller, A.H.; Teller, E. Equation of state calculations by fast computing machines. J. Chem. Phys. 1953, 21, 1087–1092

    work page 1953

  6. [5]

    Monte Carlo sampling methods using Markov chains and their applications

    Hastings, W.K. Monte Carlo sampling methods using Markov chains and their applications. Biometrika 1970, 57, 97–109

    work page 1970

  7. [6]

    Nested sampling methods

    Buchner, J. Nested sampling methods. Statist. Surv. 2023, 17, 169–215. https://doi.org/10.1214/23-ss144

    doi:10.1214/23-ss144 2023

  8. [7]

    A comparison of Bayesian and frequentist confidence intervals in the presence of a late Universe degeneracy

    Colgáin, E.O.; Pourojaghi, S.; Sheikh-Jabbari, M.M.; Sherwin, D. A comparison of Bayesian and frequentist confidence intervals in the presence of a late Universe degeneracy. arXiv 2023, arXiv:2307.16349

    arXiv 2023

Show all 43 references
  1. [9]

    Hybrid Monte Carlo

    Duane, S.; Kennedy, A.D.; Pendleton, B.J.; Roweth, D. Hybrid Monte Carlo. Phys. Lett. B 1987, 195, 216–222

    1987

  2. [10]

    Slice sampling

    Neal, R.M. Slice sampling. Ann. Stat. 2003, 31, 705–767

    2003

  3. [11]

    Nested sampling for general Bayesian computation

    Skilling, J. Nested sampling for general Bayesian computation. Bayesian Anal. 2006, 1, 833–859

    2006

  4. [12]

    PolyChord: Nested sampling for cosmology

    Handley, W.J.; Hobson, M.P .; Lasenby, A.N. PolyChord: Nested sampling for cosmology. Mon. Not. Roy. Astron. Soc. 2015, 450, L61–L65. https://doi.org/10.1093/mnrasl/slv047

    2015 doi

  5. [13]

    polychord: Next-generation nested sampling

    Handley, W.J.; Hobson, M.P .; Lasenby, A.N. polychord: Next-generation nested sampling. Mon. Not. Roy. Astron. Soc. 2015, 453, 4385–4399. https://doi.org/10.1093/mnras/stv1911

    2015 doi

  6. [14]

    PyMC: A Modern and Comprehensive Probabilistic Programming Framework in Python

    Abril-Pla, O.; Andreani, V .; Carroll, C.; Dong, L.; Fonnesbeck, C.J.; Kochurov, M.; Kumar, R.; Lao, J.; Luhmann, C.C.; Martin, O.A.; et al. PyMC: A Modern and Comprehensive Probabilistic Programming Framework in Python. PeerJ Comput. Sci. 2023, 9, e1516, https://doi.org/10.77...

    2023 doi

  7. [15]

    Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro

    Phan, D.; Pradhan, N.; Jankowiak, M. Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro. arXiv 2019, arXiv:1912.11554

    2019 arXiv

  8. [16]

    The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo

    Hoffman, M.D.; Gelman, A. The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo. J. Mach. Learn. Res. 2014, 15, 1593–1623. arXiv 2011, arXiv:1111.4246

    2014 arXiv

  9. [17]

    emcee: The MCMC Hammer.Publ

    Foreman-Mackey, D.; Hogg, D.W.; Lang, D.; Goodman, J. emcee: The MCMC Hammer.Publ. Astron. Soc. Pac. 2013, 125, 306–312. https://doi.org/10.1086/670067

    2013 doi

  10. [18]

    dynesty: A dynamic nested sampling package for estimating Bayesian posteriors and evidences

    Speagle, J.S. dynesty: A dynamic nested sampling package for estimating Bayesian posteriors and evidences. Mon. Not. Roy. Astron. Soc. 2020, 493, 3132–3158. https://doi.org/10.1093/mnras/staa278

    2020 doi

  11. [19]

    dynesty: V2.1.4, jun 2024., https://doi.org/10.5281/zenodo.12537467 accessed at 16.01.2025

    Koposov, S.; Speagle, J.; Barbary, K.; Ashton, G.; Bennett, E.; Buchner, J.; Scheffler, C.; Cook, B.; Talbot, C.; Guillochon, J.; et al. dynesty: V2.1.4, jun 2024., https://doi.org/10.5281/zenodo.12537467 accessed at 16.01.2025

    2024 doi

  12. [20]

    MultiNest: An efficient and robust Bayesian inference tool for cosmology and particle physics

    Feroz, F.; Hobson, M.P .; Bridges, M. MultiNest: An efficient and robust Bayesian inference tool for cosmology and particle physics. Mon. Not. Roy. Astron. Soc. 2009, 398, 1601–1614. https://doi.org/10.1111/j.1365-2966.2009.14548.x

    2009

  13. [21]

    Cosmology intertwined: A review of the particle physics, astrophysics, and cosmology associated with the cosmological tensions and anomalies

    Abdalla, E.; Abellan, G.F.; Aboubrahim, A.; Agnello, A.; Akarsu, O.; Akrami, Y.; Alestas, G.; Aloni, D.; Amendola, L.; Anchordoqui, L.A.; et al. Cosmology intertwined: A review of the particle physics, astrophysics, and cosmology associated with the cosmological tensions and a...

    2022 doi

  14. [22]

    Unveiling lens light complexity with a novel multi-Gaussian expansion approach for strong gravitational lensing

    He, Q.; Nightingale, J.W.; Amvrosiadis, A.; Robertson, A.; Cole, S.; Frenk, C.S.; Massey, R.; Li, R.; Cao, X.; Lange, S.C.; et al. Unveiling lens light complexity with a novel multi-Gaussian expansion approach for strong gravitational lensing. Mon. Not. Roy. Astron. Soc. 2024,...

    2024 doi

  15. [23]

    Constraining neutrino masses with weak-lensing multiscale peak counts

    Ajani, V .; Peel, A.; Pettorino, V .; Starck, J.; Li, Z.; Liu, J. Constraining neutrino masses with weak-lensing multiscale peak counts. Phys. Rev. D . 2020, 102, 103531

    2020

  16. [24]

    Utilizing Gaussian mixture models in all-sky searches for short-duration gravitational wave bursts

    Lopez, D.; Gayathri, V .; Pai, A.; Heng, I.S.; Messenger, C.; Gupta, S.K. Utilizing Gaussian mixture models in all-sky searches for short-duration gravitational wave bursts. Phys. Rev. D 2022, 105, 063024. https://doi.org/10.1103/PhysRevD.105.063024

    2022 doi

  17. [25]

    Adame, A.G. et al. [DESI Collaboration] DESI 2024 VI: Cosmological Constraints from the Measurements of Baryon Acoustic Oscillations. arXiv 2024, arXiv:2404.03002

    2024 arXiv

  18. [26]

    Constraining the dark energy models using baryon acoustic oscillations: An approach independent of H0 · rd

    Staicova, D.; Benisty, D. Constraining the dark energy models using baryon acoustic oscillations: An approach independent of H0 · rd. Astron. Astrophys. 2022, 668, A135. https://doi.org/10.1051/0004-6361/202244366

    2022 doi

  19. [27]

    Late-time constraints on interacting dark energy: Analysis independent of H0, rd, and MB

    Benisty, D.; Pan, S.; Staicova, D.; Di Valentino, E.; Nunes, R.C. Late-time constraints on interacting dark energy: Analysis independent of H0, rd, and MB. Astron. Astrophys. 2024, 688, A156. https://doi.org/10.1051/0004-6361/202449883

    2024 doi

  20. [28]

    Analytic methods for cosmological likelihoods

    Taylor, A.; Kitching, T. Analytic methods for cosmological likelihoods. Mon. Not. R. Astron. Soc. . 2010, 408, 865–875

    2010

  21. [29]

    On the effect of the degeneracy among dark energy parameters

    Gong, Y.; Gao, Q. On the effect of the degeneracy among dark energy parameters. Eur. Phys. J. . 2014, 74, 2729

    2014

  22. [30]

    Accelerating universes with scaling dark matter

    Chevallier, M.; Polarski, D. Accelerating universes with scaling dark matter. Int. J. Mod. Phys. D . 2001, 10, 213–224. Universe 2025, 1, 0 16 of 16

    2001

  23. [31]

    How many dark energy parameters?Phys

    Linder, E.V .; Huterer, D. How many dark energy parameters?Phys. Rev. D 2005, 72, 043509. https://doi.org/10.1103/PhysRevD. 72.043509

    2005 doi

  24. [32]

    Neural sampling machine with stochastic synapse allows brain-like learning and inference

    Dutta, S.; Detorakis, G.; Khanna, A.; Grisafe, B.; Neftci, E.; Datta, S. Neural sampling machine with stochastic synapse allows brain-like learning and inference. Nat. Commun. 2022, 13, 2571

    2022

  25. [33]

    Fast likelihood-free cosmology with neural density estimators and active learning

    Alsing, J.; Charnock, T.; Feeney, S.; Wandelt, B. Fast likelihood-free cosmology with neural density estimators and active learning. Mon. Not. R. Astron. Soc. 2019, 488, 4440–4458

    2019

  26. [34]

    Likelihood-free inference with neural compression of DES SV weak lensing map statistics

    Jeffrey, N.; Alsing, J.; Lanusse, F. Likelihood-free inference with neural compression of DES SV weak lensing map statistics. Mon. Not. R. Astron. Soc. 2021, 501, 954–969

    2021

  27. [35]

    A parallel tempering algorithm for probabilistic sampling and multimodal optimization

    Sambridge, M. A parallel tempering algorithm for probabilistic sampling and multimodal optimization. Geophys. J. Int. 2014, 196, 357–374

    2014

  28. [36]

    Continuously tempered hamiltonian monte carlo

    Graham, M.M.; Storkey, A.J. Continuously tempered hamiltonian monte carlo. arXiv 2017, arXiv:1704.03338

    2017 arXiv

  29. [37]

    Approach to ergodicity in Monte Carlo simulations

    Neirotti, J.; Freeman, D.L.; Doll, J. Approach to ergodicity in Monte Carlo simulations. Phys. Rev. E 2000, 62, 7445

    2000

  30. [38]

    Reducing quasi-ergodic behavior in Monte Carlo simulations by J-walking: Applications to atomic clusters

    Frantz, D.; Freeman, D.L.; Doll, J.D. Reducing quasi-ergodic behavior in Monte Carlo simulations by J-walking: Applications to atomic clusters. J. Chem. Phys. 1990, 93, 2769–2784

    1990

  31. [39]

    Orbital mcmc 2022

    Neklyudov, K.; Welling, M. Orbital mcmc 2022. Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, PMLR 151 pp. 5790–5814

    2022

  32. [40]

    Hamiltonian Monte Carlo for hierarchical models

    Betancourt, M.; Girolami, M. Hamiltonian Monte Carlo for hierarchical models. Curr. T rends Bayesian Methodol. Appl. 2015, 79, 2–4

    2015

  33. [41]

    A geometric theory of higher-order automatic differentiation

    Betancourt, M. A geometric theory of higher-order automatic differentiation. arXiv 2018, arXiv:1812.11592

    2018 arXiv

  34. [42]

    Adaptive Monte Carlo augmented with normalizing flows

    Gabrié, M.; Rotskoff, G.M.; Vanden-Eijnden, E. Adaptive Monte Carlo augmented with normalizing flows. Proc. Natl. Acad. Sci. USA 2022, 119, e2109420119

    2022

  35. [43]

    Marginal Likelihoods from Monte Carlo Markov Chains

    Heavens, A.; Fantaye, Y.; Mootoovaloo, A.; Eggers, H.; Hosenie, Z.; Kroon, S.; Sellentin, E. Marginal Likelihoods from Monte Carlo Markov Chains. arXiv 2017, arXiv:1704.03472. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are sole...

    2017 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.