Pith. sign in

REVIEW 2 major objections 6 minor 50 references

Solving High-dimensional Inverse Problems Using Amortized Likelihood-free Inference with Noisy and Incomplete Data

T0 review · 2 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A likelihood-free normalizing-flow pair estimates a 706-dimensional groundwater conductivity posterior in about two seconds per dataset, matching the accuracy of an iterative ensemble smoother benchmark without likelihood evaluations.

desk verdict Competent engineering application of known normalizing-flow components to a 706-D groundwater problem; the posterior-accuracy claim is under-supported by the evidence. read the letter →

arxiv 2412.04565 v2 pith:P7HMNNAO submitted 2024-12-05 cs.LG

Jice Zeng , Yuanzhe Wang , Alexandre M. Tartakovsky , David Barajas-Solano This is my paper · ORCID
classification cs.LG
keywords inverseproblemsnormalizingflowslikelihood-freeinferenceamortizedBayesianconditionalneuralsplineflowgroundwaterhydrologyhigh-dimensionalparameterestimation
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

This paper tries to establish that a likelihood-free, amortized Bayesian inversion method can estimate a high-dimensional heterogeneous parameter field from sparse, noisy time-series observations without ever evaluating a likelihood. Two networks are trained jointly on synthetic simulator data: a one-dimensional convolutional summary network compresses raw observations into a fixed-size feature vector, and a normalizing-flow inference network turns a latent Gaussian into posterior samples conditional on those features. In a synthetic groundwater test with 706 log-conductivity parameters and head observations at 13 sensors over 25 time steps, the flow-based posterior is comparable to the likelihood-based iterative ensemble smoother PEST-IES on relative error, relative $\ell^2$ error, and log predictive probability. Once training is done, a new measurement set is inverted in about 2 seconds, compared with about 2.5 hours for PEST-IES. The flow's predictive intervals are wider than PEST-IES's, which the paper attributes to approximation error of the trained networks, including a possibly insufficient 10-layer inference network.

What carries the argument

The central object is a conditional normalizing flow made of ten invertible layers—five conditional affine coupling layers (cACLs) and five conditional rational-quadratic spline layers (cSLs with 16 bins), stacked alternately—conditioned on a 256-dimensional feature vector produced by a 1D convolutional summary network. The affine coupling layers split the input, apply element-wise scaling and translation, and give triangular Jacobians with cheap determinants; the spline layers apply flexible monotonic element-wise maps between knots. The joint training objective is the negative log-likelihood in Eq. 22, which matches the KL divergence between the approximate and true posterior, and sampling is performed by drawing $z$ from a standard Gaussian and applying the inverse transformation. The two-network design makes the summary features themselves learned end-to-end for the inference task.

What would settle it

Take a test log-conductivity field not used in training, generate its head observations, and compare the flow's posterior to a gold-standard posterior computed by long-run MCMC or by an independent ensemble method with many more samples; if the flow's 95% credibility intervals cover the true field in far fewer than 95% of the 706 parameters, or if doubling the number of flow layers changes posterior means by more than the reported relative errors, the central claim of accurate amortized posterior estimation fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that a conditional normalizing flow, trained jointly with a learnable summary network on synthetic input–output pairs, provides an accurate amortized approximation of the posterior distribution of model parameters conditioned on noisy and incomplete data. In the demonstrated problem the parameters are the 706 log-conductivity values of a groundwater model and the data are hydraulic-head time series at 13 locations; the trained flow maps a latent Gaussian to posterior samples for any such dataset in the forward pass. Comparison with PEST-IES on two test conductivity fields shows the flow's posterior means and credibility intervals are close to the benchmark in relative error, relative $\ell^2$ error, and log predictive probability, and the flow achieves coverage of the true field that is comparable or slightly higher on the harder test case. The method also handles observation sequences with fewer time steps than in training without retraining, because the summary network compresses variable-length sequences.

Load-bearing premise

The load-bearing premise is that a 10-layer normalizing flow (five affine-coupling and five spline layers with 16 bins) trained on 4,800 synthetic examples is expressive and optimized enough to represent the true 706-dimensional posterior; the paper itself notes in Section 3.5.3 that this depth may be inadequate for 706 dimensions.

Editorial extensions

If this is right

  • A trained model converts a new observation dataset into posterior samples in about two seconds, making repeated inversion practical for monitoring or control settings where PEST-IES would require roughly 2.5 hours per dataset.
  • The model can condition on observation sequences with fewer time steps than in training (tested between 20 and 25), so a single training run covers variable-length data from the same sensor layout.
  • Because inference never evaluates the likelihood or reruns the forward model, its online cost is independent of the number of MCMC samples or ensemble iterations.
  • On the two test fields, the flow posterior is statistically close to the PEST-IES posterior on the reported accuracy metrics, and its coverage of the true field is comparable or slightly better on the more variable field.
  • The flow's posterior predictive uncertainty is larger than PEST-IES's, meaning the trained networks' approximation error is carried into predictive intervals rather than hidden.

Reading between the lines

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

  • A stress test the paper does not run is comparing the flow posterior against a long-running MCMC chain on one test field; if the flow's 95% intervals do not match the MCMC intervals in most of the 706 parameters, the learned map is biased despite agreeing with PEST-IES on aggregate metrics.
  • Scaling the inference network beyond ten layers or increasing the number of spline bins would directly test the paper's own concern that the current depth may be inadequate for 706 dimensions.
  • The architecture should transfer to other high-dimensional spatially heterogeneous fields whenever a forward simulator can generate training data, but the fixed sensor layout requirement and the roughly 4.5-hour training phase are the real costs of amortization.
  • The summary network is only validated indirectly through downstream accuracy; an explicit comparison of learned summaries against hand-crafted sufficient statistics would reveal how much information the compression discards.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 6 minor

Summary. The paper proposes an amortized likelihood-free Bayesian inversion method combining a learned summary network (1D CNN) and a normalizing-flow inference network (alternating conditional affine coupling and conditional spline layers), trained jointly on synthetic data by minimizing the negative log-likelihood. The method is applied to a 706-dimensional log-conductivity field in a MODFLOW groundwater model, using noisy head observations at 13 sparse locations over 25 time steps. The authors compare parameter and predictive posteriors with PEST-IES on two synthetic test fields and report comparable point metrics, lower inference time (about 2 seconds per dataset after 4.5 hours of training), and the ability to handle shorter observation records without retraining.

Significance. If the posterior-accuracy claim is validated, the paper would make a useful contribution: it demonstrates that a flow-based conditional generator with a learned summary statistic can be trained for a very high-dimensional (706-D) spatial inverse problem and provide near-instantaneous posterior samples once the offline training is done. The loss derivation is standard and appears correct, and the benchmark against the likelihood-based PEST-IES method is a reasonable baseline for the groundwater test case. The paper also explicitly lists known error sources in the learned approximation. However, the current evidence does not yet substantiate the central claim that the method 'accurately estimates the parameter posterior distribution' for the 706-dimensional space, because no calibration or shape-level diagnostic is reported and only two test fields are used.

major comments (2)
  1. [Section 3.5.3 and Figs. 13, 15-16] The central claim that the proposed method accurately estimates the parameter posterior distribution and the predictive posterior is not directly tested. Table 1 reports point-estimate metrics (relative error, ℓ2 error, LPP) that are insensitive to errors in spread or shape of a 706-dimensional posterior. The coverage maps in Fig. 13 are suggestive but no numerical coverage percentages are reported, and the text concedes that test #2 coverage is below 95% for both methods. In addition, the authors themselves state in Section 3.5.3 that the 10-block inference network 'may be inadequate' for the 706-dimensional parameter space. Figures 15-16 show that the proposed predictive credibility intervals are substantially wider than those of PEST-IES, which is consistent with an over-dispersed posterior approximation whose mean is nevertheless close to the reference. To support the posterior-accuracy claim, the authors should report a simulation-based calibration diagnostic (e.g., rank statistics or expected coverage per dimension) and numerical coverage rates for both the parameter and predictive posteriors. Without such evidence, the headline claim is under-supported rather than demonstrated.
  2. [Section 3.5.2, Table 1] The comparison with PEST-IES is based on only two synthetic test fields, with a single training run and no reported uncertainty on the metrics. The differences in Table 1 are small (for test #1, relative ℓ2 error 10.02% vs. 10.08%; LPP -229.44 vs. -229.93), so statements such as 'comparable accuracy' or 'slightly better performance' are not statistically meaningful without error bars, repeated training runs, or bootstrap intervals. The authors should repeat training with multiple random seeds and report means and standard deviations (over seeds) for each metric, or otherwise increase the number of test cases, before drawing conclusions about parity or superiority relative to PEST-IES.
minor comments (6)
  1. [Section 3.5.2] There is a typo: 'PEST-IT' should be 'PEST-IES' in the sentence describing Figs. 9 and 10.
  2. [Section 3.5.2] The cross-reference 'details are presented in Sec. ??' is unresolved; the intended reference to the computational-cost paragraph should be fixed.
  3. [Section 3.5.1] Figure 8 uses R² for individual parameters, but the definition of R² (e.g., coefficient of determination between posterior means and true values over what set) is not stated; please specify the formula and the reference quantity.
  4. [Section 3.5.2] The LPP formula appears to be the negative log-density of a Gaussian with variance equal to the square of the reference value; please state the assumed predictive distribution explicitly so the metric can be interpreted.
  5. [Section 3.5.1] The paper does not say whether dropout is disabled at inference time; this should be clarified, as dropout used at test time would add stochasticity to the posterior samples.
  6. [General] No code or data repository is mentioned; providing one would improve reproducibility of the results.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the posterior and predictive claims are learned from synthetic MODFLOW data and benchmarked against the external PEST-IES method, not derived from the same fitted values.

full rationale

The paper's learning objective (Eq. 23) minimizes a maximum-likelihood loss over simulated pairs generated from the prior and MODFLOW; the trained summary and inference networks are then evaluated on held-out synthetic observations and compared with PEST-IES, an independent likelihood-based smoother. No fitted parameter is renamed as a prediction: the only fitted quantities are network weights, and posterior/predictive outputs are produced by applying the trained maps and the MODFLOW forward model to test observations. The self-citations ([13], [16], [39], [47]) supply a MODFLOW implementation, an extension for longer time series, and methodological background; none is used as a uniqueness theorem or as the justification that the learned posterior is correct. The authors' own caveat that the 10-block inference network 'may be inadequate' for 706 dimensions (Sec. 3.5.3) is a validation concern, not circularity: it suggests the accuracy claim may be under-supported, but the claim does not reduce to its inputs by construction. Training and test data sharing the same prior/simulator is an in-distribution evaluation limitation, not a derivation equivalent to its inputs. Therefore no circular step can be quoted or exhibited.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central claim rests on standard flow-based density estimation plus a set of domain assumptions about the prior, noise, and simulator. The prior hyperparameters and PEST-IES settings are not reported, and the network capacity is self-admittedly uncertain, so the reader pays for several unspecified modeling choices. No new physical entities are introduced.

free parameters (3)
  • GP prior variance and correlation length (log-conductivity)
    The zero-mean Gaussian process prior with exponential covariance is stated without hyperparameter values; posterior accuracy and coverage depend on these.
  • Network and training hyperparameters = 5 cACLs + 5 cSLs, 16 bins, dropout 0.5, lr 1e-3 (decay 0.95), batch 120, 4000 epochs
    Chosen by hand with no sensitivity study; results may change under different settings.
  • Observation noise standard deviation = 0.01
    Set for the synthetic experiments; the inference method assumes this noise level, though it is not estimated from data.
assumptions (6)
  • standard math Change-of-variables formula for normalizing flows and the KL-divergence training objective are valid for the target posterior.
    Invoked in Section 3.1, Eqs. 5-19, to derive the loss; no issue.
  • domain assumption MODFLOW is an accurate forward model for the synthetic groundwater problem, and simulated outputs are treated as ground truth.
    Section 3.5 uses MODFLOW to generate all training and test data; accuracy claims are relative to this model.
  • domain assumption The log-conductivity prior is a zero-mean Gaussian process with exponential covariance, with unspecified variance and length scale.
    Section 3.5 defines the prior but not its hyperparameters; the posterior and all evaluation metrics depend on this prior.
  • domain assumption Measurement noise is additive, zero-mean Gaussian with diagonal covariance and standard deviation 0.01, uncorrelated in time.
    Section 2.2 and Section 3.5 set the noise model; all synthetic observations are generated under it.
  • ad hoc to paper The alternating cACL/cSL stack of 10 invertible blocks is sufficiently expressive to represent the 706-dimensional posterior.
    Section 3.3 and 3.5.1 choose the architecture; Section 3.5.3 admits the depth 'may be inadequate'. No expressivity or convergence analysis is given.
  • domain assumption PEST-IES was configured fairly and comparably as a benchmark, though its settings are not reported.
    Section 3.5.2 compares against PEST-IES without giving ensemble size, iterations, or regularization; fairness of the comparison is assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Solving High-dimensional Inverse Problems Using Amortized Likelihood-free Inference with Noisy and Incomplete Data." pith.science (2026). https://pith.science/paper/P7HMNNAO

@misc{pith2026241204565,
  author       = {Pith},
  title        = {Pith review of: Solving High-dimensional Inverse Problems Using Amortized Likelihood-free Inference with Noisy and Incomplete Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P7HMNNAO}},
  note         = {Machine review of arXiv:2412.04565}
}
read the original abstract

We present a likelihood-free probabilistic inversion method based on normalizing flows for high-dimensional inverse problems. The proposed method is composed of two complementary networks: a summary network for data compression and an inference network for parameter estimation. The summary network encodes raw observations into a fixed-size vector of summary features, while the inference network generates samples of the approximate posterior distribution of the model parameters based on these summary features. The posterior samples are produced in a deep generative fashion by sampling from a latent Gaussian distribution and passing these samples through an invertible transformation. We construct this invertible transformation by sequentially alternating conditional invertible neural network and conditional neural spline flow layers. The summary and inference networks are trained simultaneously. We apply the proposed method to an inversion problem in groundwater hydrology to estimate the posterior distribution of the log-conductivity field conditioned on spatially sparse time-series observations of the system's hydraulic head responses.The conductivity field is represented with 706 degrees of freedom in the considered problem.The comparison with the likelihood-based iterative ensemble smoother PEST-IES method demonstrates that the proposed method accurately estimates the parameter posterior distribution and the observations' predictive posterior distribution at a fraction of the inference time of PEST-IES.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 44 canonical work pages

  1. [1]

    Inverse problem theory and methods for model parameter estimation

    Albert Tarantola. Inverse problem theory and methods for model parameter estimation. SIAM, 2005

    2005

  2. [2]

    An introduction to inverse problems with applications

    Francisco Duarte Moura Neto and Antˆ onio Jos´ e da Silva Neto. An introduction to inverse problems with applications . Springer Science & Business Media, 2012

    work page 2012

  3. [3]

    An inverse problem approach for elasticity imaging through vibroacoustics

    Miguel A Aguilo, Wilkins Aquino, John C Brigham, and Mostafa Fatemi. An inverse problem approach for elasticity imaging through vibroacoustics. IEEE transactions on medical imaging , 29(4):1012–1021, 2010

    work page 2010

  4. [4]

    Crop physiology cali- bration in clm

    Ilias Bilionis, Beth A Drewniak, and Emil M Constantinescu. Crop physiology cali- bration in clm. Geoscientific Model Development Discussions , 7(5):6733–6771, 2014

    work page 2014

  5. [5]

    Bayesian model updating for struc- tural dynamic applications combing differential evolution adaptive metropolis and kriging model

    Jice Zeng, Young Hoon Kim, and Shiqiang Qin. Bayesian model updating for struc- tural dynamic applications combing differential evolution adaptive metropolis and kriging model. Journal of Structural Engineering , 149(6):04023070, 2023

    work page 2023

  6. [6]

    Solving inverse problems using conditional invertible neural networks

    Govinda Anantha Padmanabha and Nicholas Zabaras. Solving inverse problems using conditional invertible neural networks. Journal of Computational Physics , 433:110194, 2021

    work page 2021

  7. [7]

    Inverse analysis of asteroseismic data: a review

    Earl P Bellinger, Sarbani Basu, and Saskia Hekker. Inverse analysis of asteroseismic data: a review. Dynamics of the Sun and Stars: Honoring the Life and Work of Michael J. Thompson , pages 171–183, 2020

    work page 2020

  8. [8]

    Learning to solve Bayesian inverse problems: An amortized variational inference approach using Gaussian and Flow guides

    Sharmila Karumuri and Ilias Bilionis. Learning to solve bayesian inverse problems: An amortized variational inference approach. arXiv preprint arXiv:2305.20004 , 2023. 29

    work page Pith review arXiv 2023

Show all 50 references
  1. [9]

    Inverse problems: From regularization to bayesian inference

    Daniela Calvetti and Erkki Somersalo. Inverse problems: From regularization to bayesian inference. Wiley Interdisciplinary Reviews: Computational Statistics , 10(3):e1427, 2018

    2018

  2. [10]

    Inverse problems: a bayesian perspective

    Andrew M Stuart. Inverse problems: a bayesian perspective. Acta numerica, 19:451– 559, 2010

    2010

  3. [11]

    The bayesian approach to inverse problems

    Masoumeh Dashti and Andrew M Stuart. The bayesian approach to inverse problems. arXiv preprint arXiv:1302.6989 , 2013

    2013 arXiv

  4. [12]

    Benchmarking simulation-based inference

    Jan-Matthis Lueckmann, Jan Boelts, David Greenberg, Pedro Goncalves, and Jakob Macke. Benchmarking simulation-based inference. In International conference on artificial intelligence and statistics , pages 343–351. PMLR, 2021

    2021

  5. [13]

    A recursive inference method based on invertible neural network for multi-level model updating using video monitoring data

    Jice Zeng, Michael D Todd, and Zhen Hu. A recursive inference method based on invertible neural network for multi-level model updating using video monitoring data. Mechanical Systems and Signal Processing , 203:110736, 2023

    2023

  6. [14]

    Time series analysis by state space methods , volume 38

    James Durbin and Siem Jan Koopman. Time series analysis by state space methods , volume 38. OUP Oxford, 2012

    2012

  7. [15]

    State-of-the-art review on bayesian inference in structural system identification and damage assess- ment

    Yong Huang, Changsong Shao, Biao Wu, James L Beck, and Hui Li. State-of-the-art review on bayesian inference in structural system identification and damage assess- ment. Advances in Structural Engineering , 22(6):1329–1351, 2019

    2019

  8. [16]

    Conditional karhunen–lo` eve regression model with basis adaptation for high-dimensional problems: Uncertainty quantification and inverse modeling

    Yu-Hong Yeung, Ramakrishna Tipireddy, David A Barajas-Solano, and Alexandre M Tartakovsky. Conditional karhunen–lo` eve regression model with basis adaptation for high-dimensional problems: Uncertainty quantification and inverse modeling. Com- puter Methods in Applied Mechanic...

    2024

  9. [17]

    A survey on high-dimensional gaussian process modeling with application to bayesian optimization

    Mickael Binois and Nathan Wycoff. A survey on high-dimensional gaussian process modeling with application to bayesian optimization. ACM Transactions on Evolution- ary Learning and Optimization , 2(2):1–26, 2022

    2022

  10. [18]

    Machine learning enabled fusion of cae data and test data for vehicle crashworthiness performance evaluation by analysis

    Jice Zeng, Guosong Li, Zhenyan Gao, Yang Li, Srinivasan Sundararajan, Saeed Bar- bat, and Zhen Hu. Machine learning enabled fusion of cae data and test data for vehicle crashworthiness performance evaluation by analysis. Structural and Multidisciplinary Optimization, 66(4):96, 2023

    2023

  11. [19]

    Dimensionality reduction and polynomial chaos acceleration of bayesian inference in inverse problems.Journal of Computational Physics, 228(6):1862–1902, 2009

    Youssef M Marzouk and Habib N Najm. Dimensionality reduction and polynomial chaos acceleration of bayesian inference in inverse problems.Journal of Computational Physics, 228(6):1862–1902, 2009

    1902

  12. [20]

    Adaptive multi-fidelity polynomial chaos approach to bayesian inference in inverse problems

    Liang Yan and Tao Zhou. Adaptive multi-fidelity polynomial chaos approach to bayesian inference in inverse problems. Journal of Computational Physics , 381:110– 128, 2019. 30

    2019

  13. [21]

    Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification

    Shaoxing Mo, Nicholas Zabaras, Xiaoqing Shi, and Jichun Wu. Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification. Water Resources Research, 55(5):3856–3881, 2019

    2019

  14. [22]

    Nett: Solv- ing inverse problems with deep neural networks.Inverse Problems, 36(6):065005, 2020

    Housen Li, Johannes Schwab, Stephan Antholzer, and Markus Haltmeier. Nett: Solv- ing inverse problems with deep neural networks.Inverse Problems, 36(6):065005, 2020

    2020

  15. [23]

    Approximate bayesian computation

    Mikael Sunn ˚ aker, Alberto Giovanni Busetto, Elina Numminen, Jukka Corander, Matthieu Foll, and Christophe Dessimoz. Approximate bayesian computation. PLoS computational biology, 9(1):e1002803, 2013

    2013

  16. [24]

    Bayesflow: Learning complex stochastic models with invertible neural networks

    Stefan T Radev, Ulf K Mertens, Andreas Voss, Lynton Ardizzone, and Ullrich K¨ othe. Bayesflow: Learning complex stochastic models with invertible neural networks. IEEE transactions on neural networks and learning systems , 33(4):1452–1466, 2020

    2020

  17. [25]

    Density estimation using deep generative neural networks

    Qiao Liu, Jiaze Xu, Rui Jiang, and Wing Hung Wong. Density estimation using deep generative neural networks. Proceedings of the National Academy of Sciences , 118(15):e2101344118, 2021

    2021

  18. [26]

    Flow-GAN: Combining maximum likelihood and adversarial learning in generative models

    Aditya Grover, Manik Dhar, and Stefano Ermon. Flow-GAN: Combining maximum likelihood and adversarial learning in generative models. Proceedings of the AAAI Conference on Artificial Intelligence , 32(1), 2018

    2018

  19. [27]

    Solution of physics-based bayesian inverse problems with deep generative priors.Computer Methods in Applied Mechanics and Engineering, 400:115428, 2022

    Dhruv V Patel, Deep Ray, and Assad A Oberai. Solution of physics-based bayesian inverse problems with deep generative priors.Computer Methods in Applied Mechanics and Engineering, 400:115428, 2022

    2022

  20. [28]

    Inversion using a new low-dimensional representation of complex binary geological media based on a deep neural network

    Eric Laloy, Romain H´ erault, John Lee, Diederik Jacques, and Niklas Linde. Inversion using a new low-dimensional representation of complex binary geological media based on a deep neural network. Advances in water resources, 110:387–405, 2017

    2017

  21. [29]

    To- wards a robust parameterization for conditioning facies models using deep variational autoencoders and ensemble smoother

    Smith W A Canchumuni, Alexandre A Emerick, and Marco Aur´ elio C Pacheco. To- wards a robust parameterization for conditioning facies models using deep variational autoencoders and ensemble smoother. Computers & Geosciences , 128:87–102, 2019

    2019

  22. [30]

    Con- ditional sampling with monotone gans: from generative models to likelihood-free in- ference

    Ricardo Baptista, Bamdad Hosseini, Nikola B Kovachki, and Youssef Marzouk. Con- ditional sampling with monotone gans: from generative models to likelihood-free in- ference. arXiv preprint arXiv:2006.06755 , 2020

    2006 arXiv

  23. [31]

    Gatsbi: Generative adversarial training for simulation-based inference

    Poornima Ramesh, Jan-Matthis Lueckmann, Jan Boelts, ´Alvaro Tejero-Cantero, David S Greenberg, Pedro J Gon¸ calves, and Jakob H Macke. Gatsbi: Generative adversarial training for simulation-based inference. arXiv preprint arXiv:2203.06481 , 2022

    2022 arXiv

  24. [32]

    Bayesian inference with generative adversarial network priors

    Dhruv Patel and Assad A Oberai. Bayesian inference with generative adversarial network priors. arXiv preprint arXiv:1907.09987 , 2019. 31

    1907 arXiv

  25. [33]

    Pros and cons of gan evaluation measures: New developments

    Ali Borji. Pros and cons of gan evaluation measures: New developments. Computer Vision and Image Understanding , 215:103329, 2022

    2022

  26. [34]

    Normalizing flows for probabilistic modeling and inference

    George Papamakarios, Eric Nalisnick, Danilo Jimenez Rezende, Shakir Mohamed, and Balaji Lakshminarayanan. Normalizing flows for probabilistic modeling and inference. Journal of Machine Learning Research , 22(57):1–64, 2021

    2021

  27. [35]

    Estimation of agent-based models using bayesian deep learning ap- proach of bayesflow

    Takashi Shiono. Estimation of agent-based models using bayesian deep learning ap- proach of bayesflow. Journal of Economic Dynamics and Control , 125:104082, 2021

    2021

  28. [36]

    Outbreakflow: Model-based bayesian inference of disease outbreak dynamics with invertible neural networks and its application to the covid-19 pandemics in germany

    Stefan T Radev, Frederik Graw, Simiao Chen, Nico T Mutters, Vanessa M Eichel, Till B¨ arnighausen, and Ullrich K¨ othe. Outbreakflow: Model-based bayesian inference of disease outbreak dynamics with invertible neural networks and its application to the covid-19 pandemics in ge...

    2021

  29. [37]

    Noise-net: determining physical properties of h ii regions reflect- ing observational uncertainties

    Da Eun Kang, Ralf S Klessen, Victor F Ksoll, Lynton Ardizzone, Ullrich Koethe, and Simon CO Glover. Noise-net: determining physical properties of h ii regions reflect- ing observational uncertainties. Monthly Notices of the Royal Astronomical Society , 520(4):4981–5001, 2023

    2023

  30. [38]

    Guided image generation with conditional invertible neural networks

    Lynton Ardizzone, Carsten L¨ uth, Jakob Kruse, Carsten Rother, and Ullrich K¨ othe. Guided image generation with conditional invertible neural networks. arXiv preprint arXiv:1907.02392, 2019

    1907 arXiv

  31. [39]

    Probabilistic damage detection using a new likelihood-free bayesian inference method

    Jice Zeng, Michael D Todd, and Zhen Hu. Probabilistic damage detection using a new likelihood-free bayesian inference method. Journal of Civil Structural Health Monitoring, 13(2):319–341, 2023

    2023

  32. [40]

    Doc- umentation for the skeletal storage, compaction, and subsidence (csub) package of modflow 6

    Joseph D Hughes, Stanley A Leake, Devin L Galloway, and Jeremy T White. Doc- umentation for the skeletal storage, compaction, and subsidence (csub) package of modflow 6. Technical report, US Geological Survey, 2022

    2022

  33. [41]

    A family of nonparametric density estimation algorithms

    Esteban G Tabak and Cristina V Turner. A family of nonparametric density estimation algorithms. Communications on Pure and Applied Mathematics , 66(2):145–164, 2013

    2013

  34. [42]

    Density estimation by dual ascent of the log-likelihood

    Esteban G Tabak and Eric Vanden-Eijnden. Density estimation by dual ascent of the log-likelihood. Communications in Mathematical Sciences , 8(1):217–233, 2010

    2010

  35. [43]

    Normalizing flows: An introduction and review of current methods

    Ivan Kobyzev, Simon JD Prince, and Marcus A Brubaker. Normalizing flows: An introduction and review of current methods. IEEE transactions on pattern analysis and machine intelligence , 43(11):3964–3979, 2020

    2020

  36. [44]

    Density estimation using real nvp

    Laurent Dinh, Jascha Sohl-Dickstein, and Samy Bengio. Density estimation using real nvp. arXiv preprint arXiv:1605.08803 , 2016

    2016 arXiv

  37. [45]

    Neural spline flows

    Conor Durkan, Artur Bekasov, Iain Murray, and George Papamakarios. Neural spline flows. Advances in neural information processing systems , 32, 2019. 32

    2019

  38. [46]

    An exercise in ground-water model calibration and prediction

    David L Freyberg. An exercise in ground-water model calibration and prediction. Groundwater, 26(3):350–360, 1988

    1988

  39. [47]

    Bayesian reduced-order deep learning surrogate model for dynamic sys- tems described by partial differential equations

    Yuanzhe Wang, Yifei Zong, James L McCreight, Joseph D Hughes, and Alexandre M Tartakovsky. Bayesian reduced-order deep learning surrogate model for dynamic sys- tems described by partial differential equations. Computer Methods in Applied Me- chanics and Engineering , 429:117147, 2024

    2024

  40. [48]

    A model-independent iterative ensemble smoother for efficient history-matching and uncertainty quantification in very high dimensions

    Jeremy T White. A model-independent iterative ensemble smoother for efficient history-matching and uncertainty quantification in very high dimensions. Environ- mental Modelling & Software , 109:191–201, 2018

    2018

  41. [49]

    Mart ´ ın Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S. Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Ian Goodfellow, Andrew Harp, Geoffrey Irving, Michael Isard, Yangqing Jia, Rafal Jozefowicz, Lukasz Kaiser, Manju...

    2015

  42. [50]

    Adam: A method for stochastic optimization

    Diederik Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In International Conference on Learning Representations (ICLR) , San Diega, CA, USA, 2015. 33

    2015

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.