DegenDetector: Symbolic Recovery of Parameter Degeneracies in Bayesian Posteriors

Chaipat Tirapongprasert; Matthew Ho

REVIEW 2 major objections 5 minor 22 references

Pipeline extracts degeneracy equations from Bayesian posteriors

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · glm-5.2

2026-07-10 01:43 UTC pith:TV7J6OXD

load-bearing objection Useful pipeline for symbolic degeneracy recovery, but missing failure-mode tests the 2 major comments →

arxiv 2607.08755 v1 pith:TV7J6OXD submitted 2026-07-09 astro-ph.IM

DegenDetector: Symbolic Recovery of Parameter Degeneracies in Bayesian Posteriors

Chaipat Tirapongprasert , Matthew Ho This is my paper

classification astro-ph.IM

keywords parameter degeneracysymbolic regressionmutual informationBayesian inferenceposterior distributioncosmological parametersPlanckCMB

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper argues that parameter degeneracies in Bayesian posteriors — the situation where observed data constrains a combination of parameters more tightly than the individual values — can be automatically discovered and expressed as human-readable equations. The method works in two stages: first, mutual information screening identifies which parameters are coupled, sidestepping the combinatorial explosion of checking all possible parameter subsets; second, alternating symbolic regression fits the degeneracy as a level-set equation of the form g1(θ1) + g2(θ2) + ... + gk(θk) = c, decomposing a k-dimensional symbolic search into k independent one-dimensional problems. The decomposition is valid under the assumption that the degeneracy manifold is separable — expressible as a sum of univariate functions. On four synthetic benchmarks with known analytical degeneracies (including a Rosenbrock banana, a trigonometric surface, and an S-curve) and on Planck 2018 CMB posterior chains, the method recovers the correct functional forms with orthogonal R-squared above 0.98. Notably, applied to Planck data without any physics input, it independently recovers the cosmological horizon-angle degeneracy Ω_m h³ ≈ const, yielding an exponent ratio of approximately 2.95, close to the theoretically expected value of 3.

Core claim

The central mechanism is the alternating optimization loop: by modeling a degeneracy as a separable level set and cycling through each component function one at a time while holding the others fixed, the method converts an intractable multivariate symbolic regression into a sequence of tractable univariate ones. Combined with mutual information screening to pre-select which parameter tuples are worth fitting, this pipeline can take raw posterior samples and return a closed-form equation describing the degeneracy — including recovering known cosmological relationships from real data without domain-specific priors.

What carries the argument

Alternating symbolic regression on separable level sets, with mutual information screening for candidate selection and an orthogonal R-squared metric for goodness of fit.

Load-bearing premise

The method assumes that any degeneracy can be written as a sum of single-parameter functions set equal to a constant — that is, the degeneracy is separable. Many physical degeneracies are multiplicative or otherwise non-separable in the native parameter space. A log-transform workaround handles some of these cases, but if a degeneracy is genuinely non-separable and cannot be linearized by such a transform, the method returns a poor fit and fails to recover the functional form

What would settle it

A posterior with a known non-separable degeneracy (e.g., one involving products or compositions of parameters that no coordinate transform can render separable) would yield a low orthogonal R-squared and no recovered equation, exposing the boundary of the method's applicability.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

Degeneracy equations discovered by this method could directly inform reparameterization, letting analysts define new parameters aligned with and orthogonal to degenerate directions for tighter, more honest constraints.
The mutual information screening step serves as a standalone diagnostic: it flags which parameters are coupled even when the subsequent symbolic regression fails on non-separable degeneracies.
Applied to simulation-based inference pipelines, the tool could act as an automated post-processing step that surfaces hidden degeneracies in neural posterior estimators, where such degeneracies are common but harder to diagnose than in traditional MCMC.
The alternating optimization strategy — decomposing a multivariate symbolic regression into sequential univariate fits — could generalize beyond degeneracy detection to any implicit surface fitting problem where approximate separability holds.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If separability is the main bottleneck, a natural extension would be iterative: fit a separable model, examine residuals for remaining structure, and re-apply the pipeline to the residuals — potentially recovering degeneracies that are sums of separable terms plus corrections.
The method's success on Planck data suggests it could serve as an automated consistency check on published cosmological results: if it independently recovers known degeneracies from posterior chains, it could also surface previously unrecognized ones in newer datasets.
The orthogonal R-squared metric defined here could become a standard diagnostic for degeneracy characterization quality, analogous to how Gelman-Rubin statistics are used for MCMC convergence — giving analysts a single number to decide whether a degeneracy has been adequately captured.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

Useful pipeline for symbolic degeneracy recovery, but missing failure-mode tests

read the letter

This paper presents DegenDetector, a pipeline that takes posterior samples and returns closed-form symbolic equations for parameter degeneracies. The core idea—use k-NN mutual information to screen candidate parameter tuples, then fit an implicit separable surface via alternating symbolic regression with PySR—is genuinely new as an integrated tool. The synthetic benchmarks have known ground truth, and the Planck 2018 application recovers the established Ω_m h³ ≈ const degeneracy without being told the physics. Code is public on GitHub. The orthogonal R² metric is a sensible choice for implicit surfaces, and the alternating optimization trick that reduces a k-dimensional fit to k one-dimensional symbolic regressions is clean and practical. Credit earned here: the pipeline works, it ships, and it addresses a real gap—corner plots show correlations but not functional forms, and high-order degeneracies don't show up in pairwise projections at all. The stress-test concern about R²_perp as a false-positive safeguard is the right thing to worry about. The paper states that non-separable degeneracies will produce low R²_perp, but every benchmark is separable or log-separable. There's no test case with a genuinely non-separable degeneracy (say, θ₁·θ₂ + θ₃ = c) to verify the diagnostic actually catches failure. A separable function can approximate a non-separable one well over a bounded region, potentially yielding a deceptively high R²_perp. This is the main soft spot, and it's real—the self-diagnostic is untested. The Planck coefficient ratio of ~2.947 vs. the expected 3 is minor; the paper notes the empirical literature finds ~3.4, and the fit is in log-space with finite samples. More bothersome is the absence of error bars on recovered coefficients and underspecified details like PySR operator sets and convergence criteria. These are fixable but matter for reproducibility. The separability assumption is acknowledged as the primary limitation, which is honest. The paper is for cosmologists and SBI practitioners who hit degeneracies and want interpretable equations rather than just corner plots. It's a methods paper with a working tool and clear validation. The missing failure-mode analysis is the one thing that genuinely weakens the central claim of operating 'without domain-specific input.' Recommend serious peer review. The referee should push for a non-separable test case and error quantification, but the approach is sound and the contribution is real.

Referee Report

2 major / 5 minor

Summary. The manuscript introduces DegenDetector, a framework for identifying parameter degeneracies in Bayesian posteriors and expressing them as closed-form symbolic equations. The pipeline proceeds in two stages: (1) mutual information (MI) screening ranks k-tuples of parameters by their aggregate pairwise MI to pinpoint candidate degenerate subsets, and (2) alternating symbolic regression fits a separable level-set model g_1(theta_j1) + ... + g_k(theta_jk) = c using PySR, cycling through one-dimensional subproblems. The method is validated on four synthetic benchmarks with known ground-truth degeneracies (S-curve, Banana, Trig, Cubic) and applied to Planck 2018 posteriors, where it recovers the CMB horizon-angle degeneracy Omega_m h^3 ~ const in log-space with R^2_perp ~ 0.98. The approach is motivated by the limitation that corner plots reveal pairwise correlations but not the functional form of multi-parameter degeneracies.

Significance. The problem addressed is real and practical: high-order degeneracies that do not manifest in pairwise projections are a known challenge in cosmological inference, and automated recovery of interpretable closed-form expressions is a useful contribution. The strengths of the manuscript include the clean algorithmic design (MI screening to avoid combinatorial explosion, followed by alternating optimization that decomposes a k-dimensional problem into k one-dimensional symbolic regressions), the use of synthetic benchmarks with pre-defined ground truth, and the application to a well-established cosmological degeneracy. The public release of code on GitHub is a positive step for reproducibility. The central claim—that the pipeline can identify degenerate parameter subsets and recover their functional form without domain-specific input—is supported by the benchmarks, though its scope is constrained by the separability assumption discussed below.

major comments (2)

[§2.2 (Orthogonal Fit Quality) and §3] The manuscript states that 'when true degeneracy is not separable, the orthogonal R^2 will return a low score, indicating that the functional form has not been captured.' This self-diagnostic property is the method's only safeguard against false positives when the separable model (Eq. 2) is misspecified, and it underpins the claim of operation 'without domain-specific input.' However, all four benchmarks (S-curve, Banana, Trig, Cubic) and the Planck application involve degeneracies that are either directly separable or log-separable. There is no test case with a genuinely non-separable degeneracy (e.g., theta_1 * theta_2 + theta_3 = c, or sin(theta_1 * theta_2) + theta_3 = c) to verify that R^2_perp actually drops as claimed. This matters because a separable function can approximate a non-separable one well over a bounded region of parameter space, potentially yielding a deceptively high
[§4, Eq. (4)] The Planck recovery yields a coefficient ratio of 123.97/42.07 ~ 2.947 rather than the expected value of 3 for the Omega_m h^3 degeneracy. The manuscript describes this as 'reasonably consistent' but provides no error bars, uncertainty quantification, or goodness-of-fit assessment on the recovered coefficients themselves. Without a measure of statistical uncertainty on the ratio, it is unclear whether the deviation from 3 is within expected noise or indicates a systematic bias in the fitting procedure. The residual standard deviation sigma_perp = 0.128 in log-space is reported, but this characterizes the scatter of samples around the fitted plane, not the uncertainty on the coefficient ratio. Adding a bootstrap or jackknife estimate of the uncertainty on the recovered exponents would substantially strengthen the quantitative claim.

minor comments (5)

[§2.1] The MI-based ranking assigns each k-tuple a score equal to the sum of all pairwise MI values among its elements. For k > 2, this pairwise-sum score may not capture higher-order interactions that only manifest jointly (e.g., three parameters that are pairwise independent but jointly dependent). The authors should briefly acknowledge this limitation of the screening step.
[Figure 1] The figure captions and axis labels are difficult to parse in places. For instance, in panel (d), the equation label appears garbled ('2 2 / 1 + 2 / 2 3 = 0.5'), and in panel (f), the label reads '1 (10 2)3 0'. These should be corrected to clearly state the ground-truth degeneracy equations.
[§2.2, Eq. (3)] The epsilon in the denominator of the orthogonal loss is introduced without comment. A brief note on its role (numerical stabilization near critical points of F) and its typical magnitude would help reproducibility.
[§3] The Trig benchmark tests whether the method can approximate periodic functions 'through Taylor expansions or Gaussian-envelope estimations.' It would be useful to state explicitly what functional form PySR actually returned for this case, since the recovered expression is not shown in the text.
[§5 (Conclusion)] The sentence 'All code is publicly available on Github' should include the URL in the main text body (it appears only in a footnote) for visibility.

Circularity Check

0 steps flagged

No circularity: the method is a discovery pipeline validated against externally defined benchmarks, not a derivation chain that reduces to its inputs.

full rationale

The paper presents DegenDetector, a pipeline that takes posterior samples as input and outputs symbolic equations describing degeneracies. The derivation chain is: (1) mutual information screening ranks parameter tuples, (2) alternating symbolic regression fits separable component functions, (3) orthogonal R^2 scores the fit. No step in this chain reduces to its inputs by construction. The benchmarks (S-curve, Banana, Trig, Cubic) use degeneracies defined a priori by the authors as test functions, not fitted constants relabeled as predictions. The Planck application targets the Omega_m h^3 ~ const degeneracy established by Percival et al. (2002) and Efstathiou & Bond (1999), external citations with no author overlap. The recovered coefficient ratio of ~2.947 is compared against the known value of ~3, constituting an independent check, not a self-fulfilling prophecy. The separability assumption (Eq. 2) is a modeling constraint, not a circularity: it is openly acknowledged as a limitation, and the R^2_perp diagnostic is designed to flag when it fails. The skeptic's concern that R^2_perp is never tested on genuinely non-separable degeneracies is a valid correctness risk, but it is not circularity: the paper does not claim R^2_perp works because of a self-citation or by definition. No self-citations are load-bearing. No fitted parameter is renamed as a prediction. No uniqueness theorem is invoked. The central claim is independently grounded against external benchmarks.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The paper introduces no new physical entities, particles, forces, or dimensions. It is a methodological paper. The free parameters are fitted constants and discovered functions, which are outputs of the method rather than ad hoc postulates. The axioms are domain assumptions about the structure of degeneracies and the reliability of standard tools.

free parameters (3)

c (level set constant)
Fitted during alternating optimization as the mean of the sum of component functions (Section 2.2).
Component functions {g_l}
Discovered via symbolic regression (PySR) during alternating optimization. These are the output of the method, not inputs, but they are fitted to the posterior samples.
Planck fit coefficients (123.97, 42.07) = 123.97, 42.07
Coefficients in Eq. (4) fitted to Planck posterior samples in log-space. The ratio ~2.947 is compared to the theoretical value of 3.

axioms (4)

domain assumption Degeneracies can be modeled as the level set of a separable function (Eq. 2).
Invoked in Section 2.2 as the functional form for symbolic surface fitting. This is the core modeling assumption; the authors acknowledge it as the primary limitation in the Conclusion.
domain assumption Mutual information ranking correctly identifies degenerate parameter tuples.
Invoked in Section 2.1. The paper assumes that higher aggregated pairwise MI implies a stronger degeneracy. This holds for the tested cases but is not proven in general.
standard math k-nearest-neighbor MI estimator (Kraskov et al. 2004) provides reliable MI estimates from posterior samples.
Standard estimator from the cited literature, used in Section 2.1.
domain assumption PySR symbolic regression can recover component functions from noisy regression targets.
Invoked implicitly in Section 2.2. The alternating optimization produces noisy targets y_l, and the paper assumes PySR can handle this noise.

pith-pipeline@v1.1.0-glm · 10017 in / 2434 out tokens · 200752 ms · 2026-07-10T01:43:40.014423+00:00 · methodology

0 comments

read the original abstract

We introduce DegenDetector, a framework for identifying and characterizing parameter degeneracies in posterior distributions as closed-form symbolic equations. By combining mutual information screening with alternating symbolic regression, we facilitate automated and interpretable identification of degenerate relationships without domain-specific input. While standard tools such as corner plots can indicate that correlations exist, they do not reveal the underlying functional form. DegenDetector fills this gap by expressing multi-parameter degeneracies as closed-form equations, providing interpretable structure that scales to high-order parameter spaces.

Figures

Figures reproduced from arXiv: 2607.08755 by Chaipat Tirapongprasert, Matthew Ho.

**Figure 1.** Figure 1: Validation of DEGENDETECTOR on four benchmarks. Top row shows the full pipeline for the S-curve degeneracy. (a) Pairwise MI matrix: the degenerate parameter tuple (x, y, z) forms a high MI cluster with pairwise MI ≈ 0.47–0.48, while nuisance parameters are cleanly separated. (b) Fitted surface (green) recovers the posterior samples with R2 ⊥ = 0.9931. (c) Two-dimensional x–z projection of the same fit. (d-… view at source ↗

**Figure 2.** Figure 2: Automated recovery of the CMB horizon-angle degeneracy from Planck 2018 posterior samples. (a) Scatter of 25,225 Planck TTTEEE+lowl+lowE+lensing samples in logparameter space, overlaid with the fitted degeneracy line (red). The tight linear locus confirms that Ωmh 3 ≈ const is the dominant constraint. (b–c) True vs. predicted values for each component function g1(H0) and g2(Ωm), demonstrating that both a… view at source ↗

discussion (0)

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages · 1 internal anchor

[1]

doi:10.1051/0004-6361/201833910 , abstract =

Aghanim, N. and others , collaboration =. , year =. doi:10.1051/0004-6361/201833910 , archivePrefix =

work page doi:10.1051/0004-6361/201833910
[2]

Abbott, T. M. C. and others , collaboration =. , year =. doi:10.1103/PhysRevD.105.023520 , archivePrefix =

work page doi:10.1103/physrevd.105.023520
[3]

Adame, A. G. and others , collaboration =. , year =. doi:10.1088/1475-7516/2025/02/021 , archivePrefix =

work page doi:10.1088/1475-7516/2025/02/021 2025
[4]

Abbott, B. P. and others , collaboration =. Observation of Gravitational Waves from a Binary. , year =. doi:10.1103/PhysRevLett.116.061102 , archivePrefix =

work page doi:10.1103/physrevlett.116.061102
[5]

Communications in Applied Mathematics and Computational Science , year =

Goodman, Jonathan and Weare, Jonathan , title =. Communications in Applied Mathematics and Computational Science , year =

work page
[6]

Bayesian Analysis , year =

Skilling, John , title =. Bayesian Analysis , year =

work page
[7]

2009, title Clustering of luminous red galaxies - III

Feroz, F. and Hobson, M. P. and Bridges, M. , title =. , year =. doi:10.1111/j.1365-2966.2009.14548.x , archivePrefix =

work page doi:10.1111/j.1365-2966.2009.14548.x 2009
[8]

, year =

Lewis, Antony and Bridle, Sarah , title =. , year =. doi:10.1103/PhysRevD.66.103511 , archivePrefix =

work page doi:10.1103/physrevd.66.103511
[9]

and Lang, Dustin and Goodman, Jonathan , date =

emcee: The. , year =. doi:10.1086/670067 , archivePrefix =

work page doi:10.1086/670067
[10]

Proceedings of the National Academy of Sciences , year =

Cranmer, Kyle and Brehmer, Johann and Louppe, Gilles , title =. Proceedings of the National Academy of Sciences , year =. doi:10.1073/pnas.1912789117 , archivePrefix =

work page doi:10.1073/pnas.1912789117
[11]

and Murray, Iain , title =

Papamakarios, George and Sterratt, David C. and Murray, Iain , title =. Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics , series =. 2019 , volume =

work page 2019
[12]

Efstathiou, George and Bond, J. R. , title =. , year =. doi:10.1046/j.1365-8711.1999.02274.x , archivePrefix =

work page doi:10.1046/j.1365-8711.1999.02274.x 1999
[13]

1997, ApJ, 484, 560, doi: 10.1086/304372 Jim´ enez, E., Padilla, N., Contreras, S., et al

Jain, Bhuvnesh and Seljak, Uro. Cosmological Model Predictions for Weak Lensing: Linear and Nonlinear Regimes , journal =. 1997 , volume =. doi:10.1086/304372 , archivePrefix =

work page doi:10.1086/304372 1997
[14]

The Journal of Open Source Software , year =

corner.py: Scatterplot matrices in. The Journal of Open Source Software , year =

work page
[15]

, title =

Jasche, Jens and Wandelt, Benjamin D. , title =. , year =. doi:10.1093/mnras/stt449 , archivePrefix =

work page doi:10.1093/mnras/stt449
[16]

Physical Review E , author =

Kraskov, Alexander and St. Estimating mutual information , journal =. 2004 , volume =. doi:10.1103/PhysRevE.69.066138 , archivePrefix =

work page internal anchor Pith review doi:10.1103/physreve.69.066138 2004
[17]

2023 , archivePrefix =

Cranmer, Miles , title =. 2023 , archivePrefix =

work page 2023
[18]

PeerJ Computer Science , year =

Meurer, Aaron and others , title =. PeerJ Computer Science , year =

work page
[19]

2025 , eprint=

A comparison of Bayesian and frequentist confidence intervals in the presence of a late Universe degeneracy , author=. 2025 , eprint=

work page 2025
[20]

Percival, W. J. and Sutherland, W. and Peacock, J. A. and Baugh, C. M. and Bland-Hawthorn, J. and Bridges, T. and Cannon, R. and Cole, S. and Colless, M. and Collins, C. and Couch, W. and Dalton, G. and De Propris, R. and Driver, S. P. and Efstathiou, G. and Ellis, R. S. and Frenk, C. S. and Glazebrook, K. and Jackson, C. and Lahav, O. and Lewis, I. and L...

work page doi:10.1046/j.1365-8711.2002.06001.x 2002
[21]

Cosmological parameter estimation and inference using deep summaries , volume=

Fluri, Janis and Kacprzak, Tomasz and Refregier, Alexandre and Lucchi, Aurelien and Hofmann, Thomas , year=. Cosmological parameter estimation and inference using deep summaries , volume=. Physical Review D , publisher=. doi:10.1103/physrevd.104.123526 , number=

work page doi:10.1103/physrevd.104.123526
[22]

2025 , eprint=

Addressing prior dependence in hierarchical Bayesian modeling for PTA data analysis II: Noise and SGWB inference through parameter decorrelation , author=. 2025 , eprint=

work page 2025

[1] [1]

doi:10.1051/0004-6361/201833910 , abstract =

Aghanim, N. and others , collaboration =. , year =. doi:10.1051/0004-6361/201833910 , archivePrefix =

work page doi:10.1051/0004-6361/201833910

[2] [2]

Abbott, T. M. C. and others , collaboration =. , year =. doi:10.1103/PhysRevD.105.023520 , archivePrefix =

work page doi:10.1103/physrevd.105.023520

[3] [3]

Adame, A. G. and others , collaboration =. , year =. doi:10.1088/1475-7516/2025/02/021 , archivePrefix =

work page doi:10.1088/1475-7516/2025/02/021 2025

[4] [4]

Abbott, B. P. and others , collaboration =. Observation of Gravitational Waves from a Binary. , year =. doi:10.1103/PhysRevLett.116.061102 , archivePrefix =

work page doi:10.1103/physrevlett.116.061102

[5] [5]

Communications in Applied Mathematics and Computational Science , year =

Goodman, Jonathan and Weare, Jonathan , title =. Communications in Applied Mathematics and Computational Science , year =

work page

[6] [6]

Bayesian Analysis , year =

Skilling, John , title =. Bayesian Analysis , year =

work page

[7] [7]

2009, title Clustering of luminous red galaxies - III

Feroz, F. and Hobson, M. P. and Bridges, M. , title =. , year =. doi:10.1111/j.1365-2966.2009.14548.x , archivePrefix =

work page doi:10.1111/j.1365-2966.2009.14548.x 2009

[8] [8]

, year =

Lewis, Antony and Bridle, Sarah , title =. , year =. doi:10.1103/PhysRevD.66.103511 , archivePrefix =

work page doi:10.1103/physrevd.66.103511

[9] [9]

and Lang, Dustin and Goodman, Jonathan , date =

emcee: The. , year =. doi:10.1086/670067 , archivePrefix =

work page doi:10.1086/670067

[10] [10]

Proceedings of the National Academy of Sciences , year =

Cranmer, Kyle and Brehmer, Johann and Louppe, Gilles , title =. Proceedings of the National Academy of Sciences , year =. doi:10.1073/pnas.1912789117 , archivePrefix =

work page doi:10.1073/pnas.1912789117

[11] [11]

and Murray, Iain , title =

Papamakarios, George and Sterratt, David C. and Murray, Iain , title =. Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics , series =. 2019 , volume =

work page 2019

[12] [12]

Efstathiou, George and Bond, J. R. , title =. , year =. doi:10.1046/j.1365-8711.1999.02274.x , archivePrefix =

work page doi:10.1046/j.1365-8711.1999.02274.x 1999

[13] [13]

1997, ApJ, 484, 560, doi: 10.1086/304372 Jim´ enez, E., Padilla, N., Contreras, S., et al

Jain, Bhuvnesh and Seljak, Uro. Cosmological Model Predictions for Weak Lensing: Linear and Nonlinear Regimes , journal =. 1997 , volume =. doi:10.1086/304372 , archivePrefix =

work page doi:10.1086/304372 1997

[14] [14]

The Journal of Open Source Software , year =

corner.py: Scatterplot matrices in. The Journal of Open Source Software , year =

work page

[15] [15]

, title =

Jasche, Jens and Wandelt, Benjamin D. , title =. , year =. doi:10.1093/mnras/stt449 , archivePrefix =

work page doi:10.1093/mnras/stt449

[16] [16]

Physical Review E , author =

Kraskov, Alexander and St. Estimating mutual information , journal =. 2004 , volume =. doi:10.1103/PhysRevE.69.066138 , archivePrefix =

work page internal anchor Pith review doi:10.1103/physreve.69.066138 2004

[17] [17]

2023 , archivePrefix =

Cranmer, Miles , title =. 2023 , archivePrefix =

work page 2023

[18] [18]

PeerJ Computer Science , year =

Meurer, Aaron and others , title =. PeerJ Computer Science , year =

work page

[19] [19]

2025 , eprint=

A comparison of Bayesian and frequentist confidence intervals in the presence of a late Universe degeneracy , author=. 2025 , eprint=

work page 2025

[20] [20]

Percival, W. J. and Sutherland, W. and Peacock, J. A. and Baugh, C. M. and Bland-Hawthorn, J. and Bridges, T. and Cannon, R. and Cole, S. and Colless, M. and Collins, C. and Couch, W. and Dalton, G. and De Propris, R. and Driver, S. P. and Efstathiou, G. and Ellis, R. S. and Frenk, C. S. and Glazebrook, K. and Jackson, C. and Lahav, O. and Lewis, I. and L...

work page doi:10.1046/j.1365-8711.2002.06001.x 2002

[21] [21]

Cosmological parameter estimation and inference using deep summaries , volume=

Fluri, Janis and Kacprzak, Tomasz and Refregier, Alexandre and Lucchi, Aurelien and Hofmann, Thomas , year=. Cosmological parameter estimation and inference using deep summaries , volume=. Physical Review D , publisher=. doi:10.1103/physrevd.104.123526 , number=

work page doi:10.1103/physrevd.104.123526

[22] [22]

2025 , eprint=

Addressing prior dependence in hierarchical Bayesian modeling for PTA data analysis II: Noise and SGWB inference through parameter decorrelation , author=. 2025 , eprint=

work page 2025