Pith. sign in

REVIEW 3 major objections 4 minor 50 references

Constraint-Guided Symbolic Regression for Data-Efficient Kinetic Model Discovery

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read PI-ADoK, a constraint-guided symbolic regression framework, recovers kinetic rate laws structurally identical to the data-generating equations from noisy concentration data using 5–7 experiments instead of 16–18.

desk verdict A useful, clearly written extension of ADoK-S whose headline data-efficiency gain is undermined by an equilibrium constraint that appears to leak ground-truth timescale information from the same simulations used for evaluation. read the letter →

arxiv 2507.02730 v1 pith:J4GEQLRH submitted 2025-07-03 cs.CE cs.SC

Miguel Ángel de Carvalho Servia , Ilya Orson Sandoval , King Kuok (Mimi) Hii , Klaus Hellgardt , Dongda Zhang , Ehecatl Antonio del Rio Chanona This is my paper · ORCID
classification cs.CEcs.SC
keywords symbolicregressionkineticmodeldiscoverygeneticprogrammingphysicalconstraintsmodel-baseddesignofexperimentsuncertaintyquantificationcatalysisAkaikeinformationcriterion
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

PI-ADoK is a symbolic-regression framework that encodes qualitative physical knowledge of a reacting system directly into a genetic-programming search, so candidate expressions that violate expected behavior are penalized rather than simply scored by fit. Across three in-silico catalytic case studies it recovers rate laws structurally identical to the data-generating equations using 5–7 experiments, compared with 16–18 for the unconstrained ADoK-S baseline. The constraint-guided concentration surrogates also yield cleaner numerical derivatives for the rate-estimation step, and a Metropolis-Hastings stage turns the final model's parameters into credible prediction intervals. The payoff, if the behavior transfers to real laboratory conditions, is a marked cut in the experimental cost of obtaining interpretable kinetic models for process design and control.

What carries the argument

A penalty-augmented genetic-programming search over closed-form symbolic expressions. Each candidate is scored by its squared error plus a weighted penalty for violating any enabled constraint, which steers the search toward models with exact initial conditions, non-negative concentrations, monotone approach to equilibrium, and a flat tail by the imposed equilibrium horizon. The workflow first fits differentiable concentration surrogates, differentiates them to estimate reaction rates, and then evolves symbolic rate laws from those rates; AIC selects among complexity-ranked candidates, and a model-based design-of-experiments loop proposes the next initial condition as the one that maximally separates the two best current models.

What would settle it

One decisive check would be to run PI-ADoK on an in-silico case where the assumed equilibrium horizon is deliberately wrong (e.g., equilibrium set at 5 h when the true system equilibrates at 100 h) or where monotonicity is wrongly asserted; if the method still recovers the true rate law in 5–7 experiments, the constraint-based savings claim would be robust, whereas if the experiment count rises to the unconstrained baseline, the claim depends on correct priors.

Watch

Extended reading notes

Core claim

The central claim is that adding four qualitative constraints to an unconstrained genetic-programming discovery pipeline — exact initial conditions, approach to equilibrium over a fixed horizon, non-negative concentrations, and monotonic increase or decrease of products and reactants — reduces the number of experiments needed while preserving exact structural recovery of the rate law. In the nitrous-oxide decomposition case, after a sixth MBDoE-designed experiment PI-ADoK selects $r^* = k_1 C_{\mathrm{N_2O}}^2/(1 + k_2 C_{\mathrm{N_2O}})$, structurally identical to the true model, where the baseline needs 18 experiments; analogous outcomes appear for toluene hydrodealkylation (7 vs 16) and a theoretical isomerization (5 vs 16). The same constraint machinery improves early-time concentration surrogates, and posterior samples from Metropolis-Hastings place the parameter modes close to the generating constants while producing predictive bands that cover the data.

Load-bearing premise

The load-bearing premise is that the qualitative physical constraints — the equilibrium horizon, monotonicity, non-negativity, and exact initial conditions — are known to be true for the real system before any data are collected, so that a wrong or missing constraint would remove the search-space narrowing that the reported experiment savings rely on.

Editorial extensions

If this is right

  • Single-reaction catalytic systems could yield their kinetic rate law in 5–7 experiments instead of 16–18, cutting the experimental budget by 56–69 percent.
  • The modular constraint set lets an experimenter enable only the physical priors they trust, so partial qualitative knowledge can still be injected into the discovery process.
  • The Metropolis-Hastings phase turns point predictions into credible intervals, giving process design and safety analysis a quantitative handle on model uncertainty.
  • The MBDoE loop supplies a stopping rule: keep adding experiments while the best and second-best candidate models still diverge, and stop when they agree.

Reading between the lines

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

  • If the constraints are learned rather than assumed — for example, estimating the equilibrium horizon from a quick preliminary run — the method could become applicable when qualitative priors are uncertain.
  • A stress test that deliberately misspecifies a constraint (wrong equilibrium horizon, or monotonicity asserted for a species that later reverses) would reveal how much of the experiment savings is tied to correct priors.
  • For multi-reaction networks the paper says separate rate laws per species would be required, so the closed-loop experiment design would need to handle coupled species dynamics; that extension is plausible but unproven.
  • Because all three case studies are in-silico with Gaussian noise, a real laboratory validation on transient or steady-state plug-flow measurements would test whether the constraint benefits survive the additional noise and systematic error of physical experiments.
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

3 major / 4 minor

Summary. The paper introduces PI-ADoK, a symbolic-regression framework for discovering kinetic rate laws from noisy concentration measurements. The method extends the authors' prior ADoK work by adding penalty-based physical constraints (exact initial conditions, equilibrium approach, non-negativity, monotonicity) to a genetic programming search, coupling this with a model-based design of experiments (MBDoE) loop for iterative data collection and Metropolis-Hastings sampling for uncertainty quantification. The framework is tested on three in-silico catalytic case studies (N2O decomposition, toluene hydrodealkylation, and a theoretical isomerization), and the paper claims that PI-ADoK recovers structurally correct rate laws using 5–7 experiments while the unconstrained ADoK-S baseline requires 16–18 experiments, corresponding to 56–69% reductions in experimental burden. The reported results also include posterior parameter distributions and credible intervals for the final models.

Significance. If the central data-efficiency claim holds, PI-ADoK would be a useful step toward automated, interpretable kinetic model discovery in data-scarce settings. The paper has several commendable features: a clear three-phase workflow, a transparent AIC-based selection rule, an explicit MBDoE loop for discriminating experiments, and a coherent uncertainty-quantification module. However, the main quantitative conclusion is currently supported only by a single, deterministic-looking set of synthetic runs in which some constraints are derived from the same in-silico data used for evaluation. The absence of repeated stochastic runs, the lack of a constraint-misspecification ablation, and the manual tuning of several hyperparameters mean that the headline 56–69% experiment-reduction figures are not yet firmly established.

major comments (3)
  1. [Section 4.1] The equilibrium constraint is load-bearing for the reported data-efficiency gains, but it is derived from the ground-truth simulation rather than from prior knowledge available before model discovery. The text states that constraints are 'derived from the in-silico data' and then specifies that concentrations 'should converge by 50 hours, so that the difference between t = 50 h and t = 60 h tends toward zero.' The dataset described in Section 3.1 only spans [0, 10] h with 15 samples, so the 50 h relaxation horizon is not observable in the data an experimenter would have. This constraint effectively encodes the system's relaxation timescale and thereby leaks information from the generating ODE into the search. Since the claimed 66.67% reduction for N2O and the comparable reductions in the other cases are attributed to constraints narrowing the search space, the comparison against ADoK-S is not a clean test of prior-knowledge-driven efficiency. The authors should either demonstrate that all constraints can be specified from genuinely independent prior knowledge, or provide a sensitivity analysis around the equilibrium horizon (e.g., 20 h, 100 h, or no equilibrium constraint) and report how the required number of experiments changes.
  2. [Table 1 and Sections 4.1–4.3] The genetic programming search and the MBDoE loop are stochastic, yet the paper reports the number of experiments for each method as a single point value (5, 6, 7 for PI-ADoK; 16, 18, 16 for ADoK-S) with no repeated runs, no variance, and no description of how the ADoK-S convergence criterion was defined. Without multiple independent runs for both methods, the 56–69% reduction could be an artifact of a single favorable seed or of a lenient stopping rule for the baseline. The authors should report the distribution of required experiments over repeated runs (including seeds), state the convergence threshold used for both methods, and, ideally, give a statistical comparison (e.g., median and interquartile range) rather than single values.
  3. [Section 2.4 and Algorithm 1] Several hyperparameters are manually tuned, and the paper does not assess their sensitivity. Section 2.4 states that the constraint penalty weights lambda_j were 'manually fine-tuned,' and Algorithm 1 describes the MH proposal sigma as 'adjusted experimentally' to achieve a 40–50% acceptance rate. The values of these hyperparameters are not reported, and no ablation is provided. Because the method's performance may depend strongly on the penalty weights and the proposal scale, the claim that PI-ADoK is a robust, turnkey framework for data-efficient discovery is not yet supported. The authors should report the tuned values and add a sensitivity study, at least for the penalty weights and the equilibrium horizon.
minor comments (4)
  1. [Section 4.1, text near Eqs. (14)–(16)] The text refers to 'Figure 2(d)' and 'Figure 2(h)' when describing the selected rate models, but the relevant panels appear to be in Figure 3 (panels d and h). The figure numbering should be corrected to avoid confusion.
  2. [Section 4.1, Eq. (15c)] The candidate rate model r3 = -k1 C_NO + k2 + C_NO mixes a rate constant with a concentration inside an additive term, which is dimensionally inconsistent. If this is indeed the AIC-selected model before the MBDoE step, the paper should comment on why such a model is physically plausible enough to seed experimental design; otherwise the selection step seems questionable.
  3. [Section 4.1, parameter estimation text] The sentence mentions solving Eq. (5) 'with the ABC and LBFGS optimization algorithms,' but ABC is not defined anywhere in the manuscript. If ABC refers to an optimization method, it should be spelled out and cited; if it is a typo, it should be corrected.
  4. [Appendix A] The supplementary information states that code is 'available upon request.' Given the paper's emphasis on reproducibility and the stochastic nature of the method, a public code repository with seeds and run scripts would substantially strengthen the contribution.

Circularity Check

1 steps flagged · score 4.0 of 10

Data-efficiency gain is partly predetermined because the equilibrium-horizon and monotonicity constraints are derived from the ground-truth in-silico system, including a 50 h relaxation time not present in the [0,10] h dataset.

  1. other [Section 3.1 (dataset definition) and Section 4.1 (constraint derivation and results)]
    "An in-silico dataset is generated with ∆t = [0, 10] h and nt = 15 samples ... we integrate a series of mathematical constraints derived from the in-silico data ... the concentrations should converge by 50 hours, so that the difference between t = 50 h and t = 60 h tends toward zero."

    The claimed 66.67% reduction in experiments is attributed to the constraints narrowing the GP search. But the equilibrium-horizon constraint is not available from the published [0,10] h dataset; it is a property of the generating ODE. The paper itself states the constraints are 'derived from the in-silico data,' so the same ground-truth simulation supplies both the test data and part of the prior knowledge encoded into the search. The monotonicity constraints are likewise inferred from the same simulated trajectories. This is not a full equation-to-equation reduction because the final rate law (Eq. 16) is still selected by GP, but the data-efficiency claim is partially circular: the baseline ADoK-S is not given the same leaked ground-truth timescale information.

full rationale

The core symbolic-regression derivation is self-contained: PI-ADoK uses GP to fit concentration surrogates, differentiates them to estimate rates, searches rate-law structures with AIC-based selection, and applies MBDoE to design additional experiments. None of these steps defines the recovered rate law in terms of the inputs by construction, so there is no full Eq.-to-Eq. circularity. The main problem is that several constraints are admitted to be 'derived from the in-silico data,' and the equilibrium constraint explicitly uses a 50 h relaxation horizon that lies outside the published [0,10] h dataset. This leaks ground-truth timescale information into the method and inflates the reported reduction in experimental burden relative to the unconstrained baseline. The self-citations to the authors' prior ADoK work are normal baseline comparisons and are not load-bearing in a circular way. Overall, the central recovery of the kinetic model has independent content, but the headline data-efficiency comparison is partially undermined by constraints that already encode properties of the target system, warranting a moderate circularity score.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central efficiency claim rests on user-chosen penalty weights, an equilibrium horizon chosen from the simulated system's timescale, MH tuning parameters, and domain assumptions that the surrogate derivatives approximate true rates and that the encoded qualitative behaviors are correct. No new physical entities are introduced.

free parameters (5)
  • Constraint penalty weights λ_j = manually tuned, values not reported
    Section 2.4 states hyperparameters were manually fine-tuned for each case; the reported data-efficiency results depend on these weights.
  • Equilibrium time horizon = 50 h
    Section 4.1: constraints require concentrations to converge by t=50 h; this value is chosen from the simulated system's timescale and is not known a priori in general.
  • MH proposal standard deviation σ = tuned to 40-50% acceptance
    Section 2.5: proposal parameters were adjusted experimentally; posterior intervals depend on this tuning.
  • MH prior covariance = not specified numerically
    Section 2.5, Algorithm 1: prior is a multivariate normal whose covariance is based on the authors' confidence in the L-BFGS mean; this is a user-specified input.
  • Maximum expression complexity cap = user-defined, not quantified
    Section 2.1 and 4.1: the GP search caps structural complexity κ; this restricts the search space and affects convergence counts.
assumptions (5)
  • domain assumption Measured concentrations can be accurately approximated by differentiable symbolic surrogates whose derivatives estimate true reaction rates.
    Section 2.1: rate estimates r(i) ≈ dot(η)(t(i)) are central to the strong formulation; if the surrogate family cannot represent the true concentration profiles, rate estimates are biased.
  • domain assumption AIC with the whole dataset is a reliable model-selection criterion in low-data kinetic discovery.
    Section 2.2: AIC is used instead of data splitting; this assumes the information-criterion ranking identifies the data-generating model among GP candidates.
  • domain assumption Qualitative constraints (monotonicity, equilibrium, non-negativity) are valid for the systems studied and known before model discovery.
    Section 4.1: constraints are "derived from the in-silico data"; the paper assumes these behaviors are general physical priors, but they are read off the same datasets used for benchmarking.
  • standard math L-BFGS with multi-start and default Scipy settings finds the global optimum of the parameter estimation problem.
    Section 2.1: parameter refinement uses L-BFGS with multi-start; no guarantee of global optimality for non-convex symbolic expressions.
  • domain assumption The Metropolis-Hastings chain has converged to the posterior after discarding burn-in samples.
    Section 2.5: convergence is mentioned as monitored, but no diagnostics (e.g., Gelman-Rubin statistics) are reported for the presented posteriors.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Constraint-Guided Symbolic Regression for Data-Efficient Kinetic Model Discovery." pith.science (2026). https://pith.science/paper/J4GEQLRH

@misc{pith2026250702730,
  author       = {Pith},
  title        = {Pith review of: Constraint-Guided Symbolic Regression for Data-Efficient Kinetic Model Discovery},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/J4GEQLRH}},
  note         = {Machine review of arXiv:2507.02730}
}
read the original abstract

The industrialization of catalytic processes hinges on the availability of reliable kinetic models for design, optimization, and control. Traditional mechanistic models demand extensive domain expertise, while many data-driven approaches often lack interpretability and fail to enforce physical consistency. To overcome these limitations, we propose the Physics-Informed Automated Discovery of Kinetics (PI-ADoK) framework. By integrating physical constraints directly into a symbolic regression approach, PI-ADoK narrows the search space and substantially reduces the number of experiments required for model convergence. Additionally, the framework incorporates a robust uncertainty quantification strategy via the Metropolis-Hastings algorithm, which propagates parameter uncertainty to yield credible prediction intervals. Benchmarking our method against conventional approaches across several catalytic case studies demonstrates that PI-ADoK not only enhances model fidelity but also lowers the experimental burden, highlighting its potential for efficient and reliable kinetic model discovery in chemical reaction engineering.

Figures

Figures reproduced from arXiv: 2507.02730 by the authors.

Figure 1
Figure 1. Conceptual overview of the Physics Informed Automated Discovery of Kinetics (PI-ADoK) framework. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Step-by-step flow of PI-ADoK, highlighting the two main tasks: estimating derivatives (red box) and generating [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the modeling workflow for the fourth experiment in the decomposition of nitrous oxide, [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Illustration of the uncertainty quantification step for the selected kinetic model for the decomposition of nitrous [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]
Figure 5
Figure 5. Figure 5: Illustration of the modeling workflow for the second experiment in the hydrodealkylation of toluene, comparing [PITH_FULL_IMAGE:figures/full_fig_p019_5.png]
Figure 6
Figure 6. Figure 6: Illustration of the uncertainty quantification step for the selected kinetic model for the hydrodealkylation [PITH_FULL_IMAGE:figures/full_fig_p020_6.png]
Figure 7
Figure 7. Figure 7: Illustration of the modeling workflow for the second experiment in the hypothetical isomerization study, [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]
Figure 8
Figure 8. Figure 8: Illustration of the uncertainty quantification step for the selected kinetic model for the hypothetical iso [PITH_FULL_IMAGE:figures/full_fig_p023_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 34 canonical work pages

  1. [1]

    R. E. Baker, J. M. Peña, J. Jayamohan, and A. Jérusalem. Mechanistic models versus machine learning, a fight worth fighting for the biological community? Biol. Lett., 14(5):20170660, May 2018. doi:10.1098/rsbl.2017.0660

    work page arXiv 2018

  2. [2]

    K. V . Gernaey. A Perspective on PSE in Fermentation Process Development and Operation.Comput. Aided Chem. Eng., pages 123–130, 2015. doi:10.1016/b978-0-444-63578-5.50016-5

    work page doi:10.1016/b978-0-444-63578-5.50016-5 2015

  3. [3]

    Jimenez del Val, J

    I. Jimenez del Val, J. M. Nagy, and C. Kontoravdi. A dynamic mathematical model for monoclonal antibody N-linked glycosylation and nucleotide sugar donor transport within a maturing Golgi apparatus. Biotechnol. Prog., 27(6):1730–1743, 2011. doi:https://doi.org/10.1002/btpr.688

    work page doi:10.1002/btpr.688 2011

  4. [4]

    P. M. Jedrzejewski, I. Jimenez Jimenez del Val, A. Constantinou, A. Dell, S. M. Haslam, K. M. Polizzi, and C. Kontoravdi. Towards Controlling the Glycoform: A Model Framework Linking Extracellular Metabolites to Antibody Glycosylation. Int. J. Mol. Sci., 15(3):4492–4522, 2014. ISSN 1422-0067. doi:10.3390/ijms15034492

    work page doi:10.3390/ijms15034492 2014

  5. [5]

    R. T. Giessmann, N. Krausch, F. Kaspar, M. N. Cruz Bournazou, A. Wagner, P. Neubauer, and M. Gimpel. Dynamic Modelling of Phosphorolytic Cleavage Catalyzed by Pyrimidine-Nucleoside Phosphorylase. Processes, 7(6), 2019. ISSN 2227-9717. doi:10.3390/pr7060380

    work page doi:10.3390/pr7060380 2019

  6. [6]

    H. R. Sant Anna, A. G. Barreto, F. W. Tavares, and M. B. de Souza. Machine learning model and opti- mization of a PSA unit for methane-nitrogen separation. Comput Chem Eng, 104:377–391, September 2017. doi:10.1016/j.compchemeng.2017.05.006

    work page doi:10.1016/j.compchemeng.2017.05.006 2017

  7. [7]

    Zhang, E

    D. Zhang, E. A. del Rio-Chanona, P. Petsagkourakis, and J. Wagner. Hybrid physics-based and data-driven modeling for bioprocess online simulation and optimization. Biotechnol. Bioeng., 116(11):2919–2930, July 2019. doi:10.1002/bit.27120

    work page doi:10.1002/bit.27120 2019

  8. [8]

    E. A. del Rio-Chanona, X. Cong, E. Bradford, D. Zhang, and K. Jing. Review of advanced physical and data- driven models for dynamic bioprocess simulation: Case study of algae–bacteria consortium wastewater treatment. Biotechnol. Bioeng., 116(2):342–353, December 2018. doi:10.1002/bit.26881

    work page doi:10.1002/bit.26881 2018

Show all 50 references
  1. [9]

    S. Y . Park, C. H. Park, D. H. Choi, J. K. Hong, and D. Y . Lee. Bioprocess digital twins of mam- malian cell culture for advanced biomanufacturing. Curr. Opin. Chem. Eng. , 33:100702, September 2021. doi:10.1016/j.coche.2021.100702

    2021

  2. [10]

    Y . Sun, W. Nathan-Roberts, T. D. Pham, E. Otte, and U. Aickelin. Multi-fidelity Gaussian Process for Biomanu- facturing Process Modeling with Small Data. arXiv:2211.14493, 2022. doi:10.48550/ARXIV .2211.14493

    2022 doi arXiv

  3. [11]

    Petsagkourakis, I

    P. Petsagkourakis, I. O. Sandoval, E. Bradford, D. Zhang, and E.A. del Rio-Chanona. Reinforce- ment learning for batch bioprocess optimization. Comput Chem Eng , 133:106649, February 2020. doi:10.1016/j.compchemeng.2019.106649

    2020

  4. [12]

    E. A. del Rio-Chanona, J. L. Wagner, H. Ali, F. Fiorelli, D. Zhang, and K. Hellgardt. Deep learning-based surrogate modeling and optimization for microalgal biofuel production and photobioreactor design. AIChE J, 65 (3):915–923, December 2018. doi:10.1002/aic.16473

    2018 doi

  5. [13]

    G. Wu, M. Á. de Carvalho Servia, and M. Mowbray. Distributional reinforcement learning for inven- tory management in multi-echelon supply chains. Digital Chemical Engineering , 6:100073, March 2023. doi:10.1016/j.dche.2022.100073

    2023

  6. [14]

    Natarajan, R

    P. Natarajan, R. Moghadam, and S. Jagannathan. Online deep neural network-based feedback control of a Lutein bioprocess. J. Process Control, 98:41–51, February 2021. doi:10.1016/j.jprocont.2020.11.011

    2021 doi

  7. [15]

    Mowbray, H

    M. Mowbray, H. Kay, S. Kay, P. Castro Caetano, A. Hicks, C. Mendoza, A. Lane, P. Martin, and D. Zhang. Probabilistic machine learning based soft-sensors for product quality prediction in batch processes. Chemometr Intell Lab Syst, 228:104616, September 2022. doi:10.1016/j.chem...

    2022

  8. [16]

    S. Kay, H. Kay, M. Mowbray, A. Lane, C. Mendoza, P. Martin, and D. Zhang. Integrating Autoencoder and Heteroscedastic Noise Neural Networks for the Batch Process Soft-Sensor Design. Ind. Eng. Chem. Res., 61(36): 13559–13569, September 2022. doi:10.1021/acs.iecr.2c01789

    2022 doi

  9. [17]

    Kadlec, B

    P. Kadlec, B. Gabrys, and S. Strandt. Data-driven Soft Sensors in the process industry. Comput Chem Eng, 33(4): 795–814, April 2009. doi:10.1016/j.compchemeng.2008.12.012. 25 Constraint-Guided Symbolic Regression for Data-Efficient Kinetic Model Discovery

    2009 doi

  10. [18]

    Haider, F.O

    C. Haider, F.O. de Franca, B. Burlacu, and G. Kronberger. Shape-constrained multi-objective genetic programming for symbolic regression. Appl. Soft Comput., 132:109855, January 2023. doi:10.1016/j.asoc.2022.109855

    2023

  11. [19]

    Z. T. Wilson and N. V . Sahinidis. The ALAMO approach to machine learning.Comput Chem Eng, 106:785–795, November 2017. doi:10.1016/j.compchemeng.2017.02.010

    2017 doi

  12. [20]

    S. L. Brunton, J. L. Proctor, and J. N. Kutz. Discovering governing equations from data by sparse iden- tification of nonlinear dynamical systems. Proc. Natl. Acad. Sci. U.S.A. , 113(15):3932–3937, March 2016. doi:10.1073/pnas.1517384113

    2016 doi

  13. [21]

    J. R. Koza. Genetic programming as a means for programming computers by natural selection. Stat Comput, 4(2), June 1994. doi:10.1007/bf00175355

    1994 doi

  14. [22]

    C. J. Taylor, M. Booth, J. A. Manson, M. J. Willis, G. Clemens, B. A. Taylor, T. W. Chamberlain, and R. A. Bourne. Rapid, automated determination of reaction models and kinetic parameters. J. Chem. Eng., 413:127017, June 2021. doi:10.1016/j.cej.2020.127017

    2021

  15. [23]

    Neumann, L

    P. Neumann, L. Cao, D. Russo, V . S. Vassiliadis, and A. A. Lapkin. A new formulation for symbolic re- gression to identify physico-chemical laws from experimental data. J. Chem. Eng., 387:123412, May 2020. doi:10.1016/j.cej.2019.123412

    2020

  16. [24]

    Forster, D

    T. Forster, D. Vázquez, M. N. Cruz-Bournazou, A. Butté, and G. Guillén-Gosálbez. Modeling of bioprocesses via MINLP-based symbolic regression of S-system formalisms. Comput Chem Eng, 170:108108, February 2023. doi:10.1016/j.compchemeng.2022.108108

    2023

  17. [25]

    Inference of differential equation models by genetic programming

    Hitoshi Iba. Inference of differential equation models by genetic programming. Inf. Sci., 178(23):4453–4468,

  18. [26]

    Nobile, Daniela Besozzi, Paolo Cazzaniga, Dario Pescini, and Giancarlo Mauri

    Marco S. Nobile, Daniela Besozzi, Paolo Cazzaniga, Dario Pescini, and Giancarlo Mauri. Reverse engineering of kinetic reaction networks by means of Cartesian Genetic Programming and Particle Swarm Optimization. In 2013 IEEE Congress on Evolutionary Computation, pages 1594–1601...

    2013

  19. [27]

    Dev, and Mario R

    Shounak Datta, Vikrant A. Dev, and Mario R. Eden. Developing non-linear rate constant qspr using deci- sion trees and multi-gene genetic programming. Comput Chem Eng , 127:150–157, 2019. ISSN 0098-1354. doi:https://doi.org/10.1016/j.compchemeng.2019.05.013

    2019 doi

  20. [28]

    Reverse engineering of biochemical equations from time-course data by means of genetic programming

    Masahiro Sugimoto, Shinichi Kikuchi, and Masaru Tomita. Reverse engineering of biochemical equations from time-course data by means of genetic programming. Biosystems, 80(2):155–164, 2005. ISSN 0303-2647. doi:https://doi.org/10.1016/j.biosystems.2004.11.003

    2005 doi

  21. [29]

    Cornforth and Hod Lipson

    Theodore W. Cornforth and Hod Lipson. Inference of hidden variables in systems of differential equations with genetic programming. Genet Program Evolvable Mach, 14(2):155–190, November 2012. ISSN 1573-7632. doi:10.1007/s10710-012-9175-4

    2012 doi

  22. [30]

    The automated discovery of kinetic rate models – methodological frameworks

    Miguel Ángel de Carvalho Servia, Ilya Orson Sandoval, King Kuok (Mimi) Hii, Klaus Hellgardt, Dongda Zhang, and Ehecatl Antonio del Rio Chanona. The automated discovery of kinetic rate models – methodological frameworks. Digit Discov, 3(5):954–968, 2024. ISSN 2635-098X. doi:10....

    2024 doi

  23. [31]

    Machine learning uncovers analytical kinetic models of bioprocesses

    Tim Forster, Daniel Vázquez, Claudio Müller, and Gonzalo Guillén-Gosálbez. Machine learning uncovers analytical kinetic models of bioprocesses. Chem. Eng. Sci., 300:120606, December 2024. ISSN 0009-2509. doi:10.1016/j.ces.2024.120606

    2024

  24. [32]

    Learning sparse nonlinear dynamics via mixed-integer optimization

    Dimitris Bertsimas and Wes Gurnee. Learning sparse nonlinear dynamics via mixed-integer optimization. Nonlinear Dyn., 111(7):6585–6604, January 2023. doi:10.1007/s11071-022-08178-9

    2023 doi

  25. [33]

    Nathan Kutz, and Bingni W

    Floris Van Van Breugel, J. Nathan Kutz, and Bingni W. Brunton. Numerical Differentiation of Noisy Data: A Unifying Multi-Objective Optimization Framework. IEEE Access, 8:196865–196877, 2020. ISSN 2169-3536. doi:10.1109/access.2020.3034077. URL http://dx.doi.org/10.1109/access....

    2020

  26. [34]

    Symbolic Regression is NP-hard

    Marco Virgolin and Solon P Pissis. Symbolic Regression is NP-hard. Trans. Mach. Learn. Res., 2022. ISSN 2835-8856. URL https://openreview.net/forum?id=LTiaPxqe2e

    2022

  27. [35]

    Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl

    Miles Cranmer. Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl. arXiv, May

  28. [36]

    D. C. Liu and J. Nocedal. On the limited memory BFGS method for large scale optimization. Math. Program., 45 (1-3):503–528, August 1989. doi:10.1007/bf01589116

    1989 doi

  29. [37]

    R. Malouf. A Comparison of Algorithms for Maximum Entropy Parameter Estimation. In Proceedings of the 6th Conference on Natural Language Learning - Volume 20, COLING-02, page 1–7, USA, 2002. Association for Computational Linguistics. doi:10.3115/1118853.1118871. 26 Constraint-...

    2002

  30. [38]

    Virtanen, R

    P. Virtanen, R. Gommers, T. E. Oliphant, M. Haberland, T. Reddy, D. Cournapeau, E. Burovski, P. Peterson, W. Weckesser, J. Bright, S. J. van der Walt, M. Brett, J. Wilson, K. J. Millman, N. Mayorov, A. R. J. Nelson, E. Jones, R. Kern, E. Larson, C. J. Carey, ˙I. Polat, Y . Fen...

    2020

  31. [39]

    Model Structure Identification, page 85–108

    Miguel Ángel de Carvalho Servia and Ehecatl Antonio del Rio Chanona. Model Structure Identification, page 85–108. Royal Society of Chemistry, December 2023. ISBN 9781837670178. doi:10.1039/bk9781837670178- 00085

    2023 doi

  32. [40]

    H. Akaike. A new look at the statistical model identification. IEEE Trans. Autom. Control., 19(6):716–723, 1974. doi:10.1109/TAC.1974.1100705

    1974

  33. [41]

    Hunter and Albey M

    William G. Hunter and Albey M. Reiner. Designs for Discriminating Between Two Rival Models. Technometrics, 7(3):307–323, August 1965. doi:10.1080/00401706.1965.10490265

    1965

  34. [42]

    Kronberger, F

    G. Kronberger, F. O. de Franca, B. Burlacu, C. Haider, and M. Kommenda. Shape-Constrained Sym- bolic Regression - Improving Extrapolation with Prior Knowledge. Evol Comput , 30(1):75–98, 2022. doi:10.1162/evco_a_00294. URL https://doi.org/10.1162/evco_a_00294

    2022 doi

  35. [43]

    Solving symbolic regression problems with formal constraints

    Iwo Bł ˛ adek and Krzysztof Krawiec. Solving symbolic regression problems with formal constraints. InProceed- ings of the Genetic and Evolutionary Computation Conference, GECCO ’19, page 977–984. ACM, July 2019. doi:10.1145/3321707.3321743

    2019

  36. [44]

    Levenspiel

    O. Levenspiel. Chemical Reaction Engineering. John Wiley & Sons, Nashville, TN, 3 edition, August 1998

    1998

  37. [45]

    Schrecker, J

    L. Schrecker, J. Dickhaut, C. Holtze, P. Staehle, M. Vranceanu, K. Hellgardt, and K. K. Hii. Discovery of unexpectedly complex reaction pathways for the Knorr pyrazole synthesis via transient flow. React. Chem. Eng., 8(1):41–46, 2023. doi:10.1039/d2re00271j

    2023 doi

  38. [46]

    Model-based design of transient flow experiments for the identification of kinetic parameters

    Conor Waldron, Arun Pankajakshan, Marco Quaglio, Enhong Cao, Federico Galvanin, and Asterios Gavriilidis. Model-based design of transient flow experiments for the identification of kinetic parameters. React. Chem. Eng., 5:112–123, 2020. doi:10.1039/C9RE00342H

    2020 doi

  39. [47]

    H. S. Fogler. Elements of chemical reaction engineering. Prentice Hall, Philadelphia, PA, 5 edition, January 2016

    2016

  40. [48]

    G. B. Marin, G. S Yablonsky, and D. Constales. Kinetics of chemical reactions: decoding complexity. John Wiley & Sons, 2019. 27

    2019

  41. [2008]

    doi:https://doi.org/10.1016/j.ins.2008.07.029

    ISSN 0020-0255. doi:https://doi.org/10.1016/j.ins.2008.07.029. Including Special Section: Genetic and Evolutionary Computing

    2008 doi

  42. [2023]

    doi:arXiv:2305.01582v3

    arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.