Pith. sign in

REVIEW 4 major objections 5 minor 30 references

Modified UNIFAC 2.0 -- A Group-Contribution Method Completed with Machine Learning

T0 review · 4 major / 5 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 Modified UNIFAC 2.0 claims to complete all group-pair interaction parameters via machine-learned matrix completion and to nearly halve the mean error on the shared data horizon.

desk verdict A useful, practical completion of modified UNIFAC, but the claim of full predictive scope rests on a low-rank prior that is never validated on the truly zero-data group pairs. read the letter →

arxiv 2412.12962 v1 pith:Y2QUPTWD submitted 2024-12-17 physics.chem-ph

Nicolas Hayer , Hans Hasse , Fabian Jirasek This is my paper · ORCID
classification physics.chem-ph
keywords modifiedUNIFACgroup-contributionmethodmatrixcompletionactivitycoefficientsexcessenthalpypair-interactionparametersmachinelearningvapor-liquidequilibrium
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 tries to establish that the main limitation of the workhorse group-contribution method modified UNIFAC, namely missing pair-interaction parameters, can be removed by machine-learned matrix completion without changing the physical model equations. It presents modified UNIFAC 2.0, trained end-to-end on more than 500,000 experimental data points for activity coefficients and excess enthalpies, yielding a complete set of parameters for all 63 main groups. On mixtures both models can describe, the new model nearly halves the mean error in predicted logarithmic activity coefficients; on mixtures the older model cannot describe at all, the new model still performs well. If the claim holds, users get a drop-in replacement parameter table that extends prediction to any mixture decomposable into the model's 125 subgroups.

What carries the argument

The machinery is matrix completion by low-rank factorization. Each pair-interaction parameter is written as a dot product of two latent feature vectors of length $K = 8$, namely $a_{mn} = \theta^a_m \cdot \beta^a_n$ and $b_{mn} = \theta^b_m \cdot \beta^b_n$. Fitting all features simultaneously under a Bayesian likelihood with a heavy-tailed Cauchy error model lets every observed data point influence every parameter, so the model imputes values for the 594 group pairs that never appear in the training data. The standard modified UNIFAC equations then generate activity coefficients and excess enthalpies from the completed parameter tables.

What would settle it

Measure activity coefficients and excess enthalpies for binary mixtures whose group pairs have no training data, roughly 594 of the 1953 group combinations, and compare the results with modified UNIFAC 2.0 predictions; if the mean absolute error on these physically measured unseen pairs substantially exceeds the model's reported error on seen pairs, the latent-space completion is filling gaps with artifacts rather than chemistry.

Watch

Extended reading notes

Core claim

The central claim is that modified UNIFAC 2.0, trained end-to-end on more than 500,000 experimental data points for activity coefficients and excess enthalpies, produces complete pair-interaction parameter tables for all 63 main groups and achieves improved accuracy compared with the latest published modified UNIFAC (Dortmund) while significantly expanding the predictive scope. On the shared comparison horizon, the paper reports that the mean of the mixture-wise mean absolute error in logarithmic activity coefficients is nearly halved, and that on mixtures inaccessible to the older model the new model still reaches accuracy comparable to what the older model achieves where it applies. The temperature dependence is kept through the $a_{mn}$ and $b_{mn}$ parameters, while the rarely fitted $c_{mn}$ parameter is dropped from the model.

Load-bearing premise

The load-bearing premise is that each missing pair-interaction parameter can be recovered from the dot product of two 8-dimensional latent feature vectors learned from other data; if group-pair interactions are not approximately low-rank in that space, the imputed parameters for the 594 unsupported group pairs are unconstrained artifacts.

Editorial extensions

If this is right

  • Any binary or multicomponent mixture decomposable into the 125 subgroups can be passed through the model, because no missing group pair blocks the calculation.
  • Existing process simulators need only swap in the supplied complete parameter tables to gain the expanded predictive scope.
  • The mean mixture-wise mean absolute error in logarithmic activity coefficients on the shared horizon is nearly halved relative to the older model, with similar improvement seen for excess enthalpies.
  • Withheld-component and withheld-pair tests show only modest accuracy loss, so predictions for chemistries not directly fitted are plausible.
  • The training scheme can be rerun as new experimental data arrive, making the model updatable without changing its equations.

Reading between the lines

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

  • Beyond the paper, the same low-rank factorization could be applied to other group-contribution frameworks, because it treats the parameter matrix rather than the chemistry as the object to be completed.
  • A testable extension is to interpret the learned eight-dimensional latent vectors as coarse descriptors of group chemistry, checking whether distances in that space correlate with known functional-group similarity.
  • Because the Bayesian training yields posterior uncertainty for the features, one could flag group pairs whose imputed parameters carry high variance, guiding which new experiments would most reduce prediction risk.
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

4 major / 5 minor

Summary. The paper presents modified UNIFAC 2.0, a group-contribution model that replaces the sequential parameter fitting of modified UNIFAC (Dortmund) with an end-to-end trained matrix completion approach. Pair-interaction parameters are modeled as low-rank latent feature products (Eqs. (2) and (3)), enabling simultaneous training of all a_mn and b_mn parameters and completion of the parameter table. The model is trained on more than 500,000 DDB data points for activity coefficients and excess enthalpies and is compared with the latest published modified UNIFAC version (called mod. UNIFAC 1.0). The authors report improved accuracy on the shared prediction horizon and a large expansion of applicability, and they provide complete parameter tables in the Supporting Information.

Significance. If the central claim holds, the contribution is significant and practically useful: it provides a freely available, complete parameterization of modified UNIFAC for 63 main groups, which can be dropped into existing process simulators. The paper also demonstrates a promising methodological template for embedding matrix completion into established thermodynamic group-contribution frameworks. The end-to-end training on a large experimental database, the provision of complete parameter tables as .csv files, and the inclusion of a held-out component test and a held-out pair-parameter test are concrete strengths that go beyond what is typical for parameter-fitting papers. The main scientific risk is that the completeness of the parameter table relies on a low-rank assumption that is not directly validated for the 30% of group pairs that have no training data, and the headline accuracy comparison is partly in-sample for the new model.

major comments (4)
  1. [Sec. 3.3 and Sec. S.1] The central claim of complete predictive scope rests on the parameters imputed for the 594 group combinations (30% of the matrix) for which no experimental data are available (Supporting Information, Sec. S.1). The extrapolation test in Sec. 3.3 withholds 100 group combinations, but all of these, as listed in Tables S.1 and S.2, have nonzero numbers of data points and mixtures; they test interpolation among data-supported pairs, not the fully unobserved regime. To support the completeness claim, the authors should provide additional evidence for the zero-data entries, for example a stability analysis of the imputed parameters under retraining with different random seeds, a synthetic-data recovery test for a matrix with known low-rank structure, or a cross-validation scheme that removes entire rows/columns of the parameter matrix. Without such evidence, the imputed values for the 594 unsupported pairs may be numerical artifacts, and the claim of unlimited applicability within the 125 subgroups is not established.
  2. [Figs. 2 and 3 and Sec. 3.1] The headline comparison showing improved accuracy of mod. UNIFAC 2.0 over mod. UNIFAC 1.0 is partly a comparison of an in-sample model with a model that was not trained on the evaluation data. Mod. UNIFAC 2.0 is trained on the full DDB data set, including data measured after 2016, whereas mod. UNIFAC 1.0 was published in 2016 and could not have used those data. The authors acknowledge this only by saying that it is 'reasonable to assume' the training sets are similar. This is not a substitute for a controlled evaluation: the results in Figs. 2 and 3 should be reported for a version of mod. UNIFAC 2.0 trained only on data available before 2016, or the claims should be explicitly restricted to the held-out experiments of Secs. 3.2 and 3.3, which are more convincing.
  3. [Sec. 2.1 and Eq. (1)] Mod. UNIFAC 2.0 drops the c_mn parameters that appear in the original modified UNIFAC equation (Eq. (1)). The authors justify this by stating that c_mn was fitted for only very few group pairs in mod. UNIFAC 1.0, but those few fitted c_mn values can be important for the systems they describe. The comparison on the shared horizon therefore changes the model form as well as the fitting procedure. The paper should quantify how many group pairs in mod. UNIFAC 1.0 use nonzero c_mn and discuss the effect of setting them to zero on the comparison for those pairs.
  4. [Sec. 3.3 and Tables S.1/S.2] The 100 test sets in the pair-extrapolation experiment are based on a single random selection, and many contain very few data points (e.g., one mixture with one data point for pairs 16-35, 17-28, 28-39, 33-34, 33-35, 36-40 in Table S.1). The aggregate box plots in Fig. 8 and Fig. S.6 obscure large per-pair failures, such as hE MAE of 3.86 kJ/mol for pair 18-44 and 2.25 kJ/mol for pair 5-17 in Table S.2. The authors should report the distribution of per-pair errors, relate them to the amount of held-out data, and ideally repeat the random selection several times to show that the aggregate result is robust. As it stands, the claim that 'true predictions achieve comparable accuracy' is based on a summary that is dominated by a few large test sets.
minor comments (5)
  1. [Sec. 2.1] The latent dimension K=8 is stated to have been determined in preliminary studies, but no details or sensitivity analysis are provided. A short paragraph or a reference showing how K was chosen would help the reader assess the robustness of the low-rank assumption.
  2. [Sec. 2.2 and Eq. (8)] The derivation of activity coefficients from VLE data uses the extended Raoult's law with an ideal gas phase and neglects the pressure dependence of the liquid-phase chemical potential. The authors note this limitation but do not quantify its impact for the 10 bar pressure limit. A brief estimate of the expected errors in ln gamma from these approximations would strengthen the data-preprocessing description.
  3. [Figs. 2, 3, 7, 8, S.4, S.6] The box plots are informative, but they do not show the number of mixtures per box. Since sample sizes vary strongly between the 'mod. UNIFAC 1.0 horizon' and 'mod. UNIFAC 2.0 only' sets, the numbers should be given in the captions or in the text.
  4. [Sec. 3.1] The statement that the mean of the MAE is 'nearly halved' with mod. UNIFAC 2.0 is not supported by explicit numbers. Please report the exact mean and median MAE values for both models and for both the ln gamma and hE results, either in the text or in a table.
  5. [Sec. 3.1 and Fig. 4b] The paper says mod. UNIFAC 2.0 outperforms mod. UNIFAC 1.0 for 'most group combinations' and reports 461 improved versus 267 deteriorated combinations. It should be stated what the total number of compared group combinations is, so the reader can see the fraction of cases where the new model is worse.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: mod. UNIFAC 2.0 is trained end-to-end on DDB data and tested on withheld mixtures and group pairs, so the accuracy and scope claims do not reduce to the model's inputs.

full rationale

The paper's derivation chain is not circular. Pair-interaction parameters are generated by the low-rank factorization a_mn = theta_a_m dot beta_a_n and b_mn = theta_b_m dot beta_b_n (Eqs. 2-3), and those features are learned by maximizing a likelihood that compares the UNIFAC output (Eqs. 6-7) with experimental ln gamma and hE data. The target quantities (MAEs on the mod. UNIFAC 1.0 horizon and on the 'mod. UNIFAC 2.0 only' sets) are external DDB measurements, not functions of the fitted parameters alone in a way that makes the comparison definitional. The extrapolation tests in Secs. 3.2 and 3.3 withhold whole components or whole group-pair data sets before training, so the reported accuracy is a genuine holdout evaluation rather than a re-statement of the training objective. The self-citations (refs. 11-18) supply the MCM methodology and hyperparameter choices (e.g., lambda=0.4 'as in our recent work'), but the present evidence does not rest on those citations: the same paper evaluates the model against experimental data that were excluded from training. The rank-8 factorization is an unproven structural assumption, and the Sec. 3.3 test covers only 100 group combinations that have data, leaving the 594 zero-data entries indirectly supported at best; however, this is a limitation in external validity, not circular reasoning. No equation or fitted parameter is equivalent by construction to the claimed prediction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 1 invented entities

The central result rests on UNIFAC's group-additivity physics, the low-rank factorization ansatz, DDB data quality, and the ideal-gas approximation for VLE up to 10 bar. The latent vectors are invented quantities with no physical meaning, though they are constrained by held-out predictions within this paper.

free parameters (3)
  • latent feature vectors theta_a, theta_b, beta_a, beta_b for all 63 main groups = not reported; final amn and bmn tables in SI
    Values learned by variational inference to fit more than 500,000 data points; each vector has length K=8 and enters Eqs. (2) and (3).
  • latent dimension K = 8
    Set from preliminary studies (Section 2.1); controls the capacity of the matrix completion and is not cross-validated in this paper.
  • Cauchy scale lambda = 0.4
    Chosen as in the authors' prior UNIFAC 2.0 work (ref 18); appears in Eqs. (4) and (5) and controls outlier weighting in the likelihood.
assumptions (5)
  • domain assumption The modified UNIFAC equations (refs 21,22) with fixed subgroups Rk and Qk suffice to represent activity coefficients and excess enthalpies of the studied mixtures.
    Invoked throughout Section 2.1; all predictions use these equations (Eqs. 6 and 7), so any inadequacy of the physical model transfers to mod UNIFAC 2.0.
  • ad hoc to paper Pair interaction parameters factor as amn = theta_a_m dot beta_a_n and bmn = theta_b_m dot beta_b_n with latent dimension K=8.
    Eqs. (2) and (3); this low-rank ansatz is the mechanism that creates complete tables, but it is not derived from physical constraints and is not tested for the 594 group pairs with no data.
  • domain assumption DDB data quality flags and the 10 bar pressure limit yield experimental training values that are accurate enough.
    Section 2.2; the authors exclude low-quality DDB entries and restrict VLE pressure to 10 bar, but use extended Raoult's law with ideal gas and no fugacity correction, an approximation they acknowledge in the footnote.
  • standard math Mean-field variational inference with standard-normal priors gives posterior means adequate for point predictions.
    Section 2.1; the posterior is approximated by independent normals, maximizing the ELBO with Adam; no convergence diagnostics or sensitivity checks are reported.
  • domain assumption Components in the database can be decomposed into the 63 main groups and 125 subgroups of modified UNIFAC.
    Section 2.2; mixtures not decomposable are excluded, so the scope claim only applies within this group vocabulary.
invented entities (1)
  • Latent feature vectors theta_a, theta_b, beta_a, beta_b (dimension 8) for each of 63 main groups
    purpose: Low-rank representation of all amn and bmn pair parameters, enabling matrix completion of the incomplete interaction matrix.
    The latent vectors have no direct experimental counterpart and are not released; they are internal degrees of freedom. Their only support is indirect out-of-sample prediction tests in Sections 3.2 and 3.3.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Modified UNIFAC 2.0 -- A Group-Contribution Method Completed with Machine Learning." pith.science (2026). https://pith.science/paper/Y2QUPTWD

@misc{pith2026241212962,
  author       = {Pith},
  title        = {Pith review of: Modified UNIFAC 2.0 -- A Group-Contribution Method Completed with Machine Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y2QUPTWD}},
  note         = {Machine review of arXiv:2412.12962}
}
read the original abstract

Predicting thermodynamic properties of mixtures is a cornerstone of chemical engineering, yet conventional group-contribution (GC) methods like modified UNIFAC (Dortmund) remain limited by incomplete tables of pair-interaction parameters. To address this, we present modified UNIFAC 2.0, a hybrid model that integrates a matrix completion method from machine learning into the GC framework, allowing for the simultaneous training of all pair-interaction parameters, including the prediction of parameters that cannot be fitted due to missing data. Utilizing an extensive training set of more than 500,000 experimental data for activity coefficients and excess enthalpies from the Dortmund Data Bank, modified UNIFAC 2.0 achieves improved accuracy compared to the latest published version of modified UNIFAC (Dortmund) while significantly expanding the predictive scope. Its flexible design allows updates with new experimental data or customizations for specific applications. The new model can easily be implemented in established simulation software with complete parameter tables readily available.

Figures

Figures reproduced from arXiv: 2412.12962 by the authors.

Figure 1
Figure 1. Comparison of mod. UNIFAC 1.0 5 and mod. UNIFAC 2.0 (this work). Mod. UNI￾FAC 1.0 relies on sequential parameter fitting, whereas mod. UNIFAC 2.0 integrates a matrix completion method (MCM) for predicting pair-interaction parameters into the mod. UNI￾FAC framework. Mod. UNIFAC 2.0 was trained end-to-end on experimental logarithmic activity coefficients (ln γi) and excess enthalpy (h E ) data. After training, the com… view at source ↗
Figure 2
Figure 2. Mean absolute error (MAE) of the predicted [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Mean absolute error (MAE) of the predicted [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: (a) Heatmap of the mean absolute error (MAE) of the predicted [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Prediction of ln γi and isothermal vapor–liquid phase diagrams for binary mixtures with mod. UNIFAC 2.0 (lines) and comparison to experimental data from the DDB (sym￾bols). Mod. UNIFAC 1.0 is not applicable to the mixtures shown [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Prediction of excess enthalpies h E at 298.15 K for binary mixtures with mod. UNI￾FAC 2.0 (lines) and comparison to experimental data from the DDB (symbols). Mod. UNI￾FAC 1.0 is not applicable to the mixtures shown. Furthermore, since mod. UNIFAC 2.0 is based on pairwi…
Figure 7
Figure 7. Figure 7: Mean absolute error (MAE) of the predicted [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Mean absolute error (MAE) of the predicted [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 26 canonical work pages

  1. [1]

    L.; Prausnitz, J

    Fredenslund, A.; Jones, R. L.; Prausnitz, J. M. Group-contribution estimation of activity coefficients in nonideal liquid mixtures. AIChE Journal 1975, 21, 1086--1099

    work page 1975

  2. [2]

    UNIFAC parameter table for prediction of liquid-liquid equilibriums

    Magnussen, T.; Rasmussen, P.; Fredenslund, A. UNIFAC parameter table for prediction of liquid-liquid equilibriums. Industrial & Engineering Chemistry Process Design and Development 1981, 20, 331--339

    work page 1981

  3. [3]

    Vapor - Liquid Equilibria by UNIFAC Group Contribution

    Wittig, R.; Lohmann, J.; Gmehling, J. Vapor - Liquid Equilibria by UNIFAC Group Contribution. 6. Revision and Extension. Industrial & Engineering Chemistry Research 2003, 42, 183--188

    work page 2003

  4. [4]

    2023; http://www.unifac.org

    The UNIFAC Consortium. 2023; http://www.unifac.org

    work page 2023

  5. [5]

    Further Development of Modified UNIFAC (Dortmund): Revision and Extension 6

    Constantinescu, D.; Gmehling, J. Further Development of Modified UNIFAC (Dortmund): Revision and Extension 6. Journal of Chemical & Engineering Data 2016, 61, 2738--2748

    work page 2016

  6. [6]

    2024; www.ddbst.com

    DDBST - Dortmund Data Bank Software & Separation Technology GmbH Dortmund Data Bank. 2024; www.ddbst.com

    work page 2024

  7. [7]

    CHEMCAD V.8

    Chemstations, Inc. CHEMCAD V.8. 2024; www.chemstations.com

    work page 2024

  8. [8]

    Aspen Plus V14.5

    Aspen Technology, Inc. Aspen Plus V14.5. 2024; www.aspentech.com

    work page 2024

Show all 30 references
  1. [9]

    Ramlatchan ; M

    A. Ramlatchan ; M. Yang ; Q. Liu ; M. Li ; J. Wang ; Y. Li A survey of matrix completion methods for recommendation systems. Big Data Mining and Analytics 2018, 1, 308--323

    2018

  2. [10]

    Matrix Factorization Techniques for Recommender Systems

    Koren, Y.; Bell, R.; Volinsky, C. Matrix Factorization Techniques for Recommender Systems. Computer 2009, 42, 30--37

    2009

  3. [11]

    Hybridizing physical and data-driven prediction methods for physicochemical properties

    Jirasek, F.; Bamler, R.; Mandt, S. Hybridizing physical and data-driven prediction methods for physicochemical properties. Chemical Communications 2020, 56, 12407--12410

    2020

  4. [12]

    Jirasek, F.; Alves, R. A. S.; Damay, J.; Vandermeulen, R. A.; Bamler, R.; Bortz, M.; Mandt, S.; Kloft, M.; Hasse, H. Machine Learning in Thermodynamics: Prediction of Activity Coefficients by Matrix Completion. The journal of physical chemistry letters 2020, 11, 981--985

    2020

  5. [13]

    Predicting Activity Coefficients at Infinite Dilution for Varying Temperatures by Matrix Completion

    Damay, J.; Jirasek, F.; Kloft, M.; Bortz, M.; Hasse, H. Predicting Activity Coefficients at Infinite Dilution for Varying Temperatures by Matrix Completion. Industrial & Engineering Chemistry Research 2021, 60, 14564--14578

    2021

  6. [14]

    Prediction of Henry's law constants by matrix completion

    Hayer, N.; Jirasek, F.; Hasse, H. Prediction of Henry's law constants by matrix completion. AIChE Journal 2022, 68, e17753

    2022

  7. [15]

    Database for liquid phase diffusion coefficients at infinite dilution at 298 K and matrix completion methods for their prediction

    Gro mann, O.; Bellaire, D.; Hayer, N.; Jirasek, F.; Hasse, H. Database for liquid phase diffusion coefficients at infinite dilution at 298 K and matrix completion methods for their prediction. Digital Discovery 2022, 1, 886--897

    2022

  8. [16]

    Making thermodynamic models of mixtures predictive by machine learning: matrix completion of pair interactions

    Jirasek, F.; Bamler, R.; Fellenz, S.; Bortz, M.; Kloft, M.; Mandt, S.; Hasse, H. Making thermodynamic models of mixtures predictive by machine learning: matrix completion of pair interactions. Chemical science 2022, 13, 4854--4862

    2022

  9. [17]

    Prediction of parameters of group contribution models of mixtures by matrix completion

    Jirasek, F.; Hayer, N.; Abbas, R.; Schmid, B.; Hasse, H. Prediction of parameters of group contribution models of mixtures by matrix completion. Physical chemistry chemical physics : PCCP 2023, 25, 1054--1062

    2023

  10. [18]

    Advancing Thermodynamic Group-Contribution Methods by Machine Learning: UNIFAC 2.0

    Hayer, N.; Wendel, T.; Mandt, S.; Hasse, H.; Jirasek, F. Advancing Thermodynamic Group-Contribution Methods by Machine Learning: UNIFAC 2.0. http://arxiv.org/pdf/2408.05220

    arXiv

  11. [19]

    2023; www.ddbst.com

    DDBST - Dortmund Data Bank Software & Separation Technology GmbH Dortmund Data Bank. 2023; www.ddbst.com

    2023

  12. [20]

    Gmehling, J.; Kolbe, B.; Kleiber, M.; Rarey, J. R. Chemical thermodynamics for process simulation; Wiley-VCH-Verl.: Weinheim, 2012

    2012

  13. [21]

    A modified UNIFAC model

    Weidlich, U.; Gmehling, J. A modified UNIFAC model. 1. Prediction of VLE, hE, and ^ . Industrial & Engineering Chemistry Research 1987, 26, 1372--1381

    1987

  14. [22]

    A modified UNIFAC model

    Gmehling, J.; Li, J.; Schiller, M. A modified UNIFAC model. 2. Present parameter matrix and results for different thermodynamic properties. Industrial & Engineering Chemistry Research 1993, 32, 178--193

    1993

  15. [23]

    P.; Jankowiak, M.; Obermeyer, F.; Pradhan, N.; Karaletsos, T.; Singh, R.; Szerlip, Paul and Horsfall, Paul ; Goodman, N

    Bingham, E.; Chen, J. P.; Jankowiak, M.; Obermeyer, F.; Pradhan, N.; Karaletsos, T.; Singh, R.; Szerlip, Paul and Horsfall, Paul ; Goodman, N. D. Pyro: Deep Universal Probabilistic Programming. Journal of Machine Learning Research 2018,

    2018

  16. [24]

    M.; Kucukelbir, A.; McAuliffe, J

    Blei, D. M.; Kucukelbir, A.; McAuliffe, J. D. Variational Inference: A Review for Statisticians. Journal of the American Statistical Association 2017, 112, 859--877

    2017

  17. [25]

    P.; Ba, J

    Kingma, D. P.; Ba, J. Adam: A Method for Stochastic Optimization. http://arxiv.org/pdf/1412.6980.pdf

    arXiv

  18. [26]

    Conductor-like Screening Model for Real Solvents: A New Approach to the Quantitative Calculation of Solvation Phenomena

    Klamt, A. Conductor-like Screening Model for Real Solvents: A New Approach to the Quantitative Calculation of Solvation Phenomena. The Journal of Physical Chemistry 1995, 99, 2224--2235

    1995

  19. [27]

    COSMO-RS: a novel and efficient method for the a priori prediction of thermophysical data of liquids

    Klamt, A.; Eckert, F. COSMO-RS: a novel and efficient method for the a priori prediction of thermophysical data of liquids. Fluid Phase Equilibria 2000, 172, 43--72

    2000

  20. [28]

    COSMO-RS: From quantum chemistry to fluid phase thermodynamics and drug design, 1st ed.; Elsevier: Amsterdam, 2005

    Klamt, A. COSMO-RS: From quantum chemistry to fluid phase thermodynamics and drug design, 1st ed.; Elsevier: Amsterdam, 2005

    2005

  21. [29]

    mod. UNIFAC 2.0 only

    Hsieh, C.-M.; Lin, S.-T.; Vrabec, J. Considering the dispersive interactions in the COSMO-SAC model for more accurate predictions of fluid phase behavior. Fluid Phase Equilibria 2014, 367, 109--116 mcitethebibliography document 2_ESI.tex0000664000000000000000000006722614730306...

    2014

  22. [30]

    H`R ,.e

    The UNIFAC Consortium. 2023; http://www.unifac.org mcitethebibliography document anc/0000775000000000000000000000000014730306662010326 5ustar rootrootanc/a_Parameters_modUNIFAC20.csv0000664000000000000000000011566514730306662015353 0ustar rootroot;1;2;3;4;5;6;7;8;9;10;11;12;13...

    2023

Pith tools

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