Pith. sign in

REVIEW 4 major objections 6 minor 43 references

On the Implementation of a Bayesian Optimization Framework for Interconnected Systems

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read BOIS uses an adaptive linearization of a known composite function to obtain closed-form estimates of performance mean and uncertainty, letting Bayesian optimization exploit structure instead of sampling.

desk verdict A well-executed implementation paper whose central performance claim is plausible but not yet proven: the MC-BO baseline is underpowered and the reported gaps lack significance tests. read the letter →

arxiv 2501.00967 v1 pith:WWL2ZQE3 submitted 2025-01-01 stat.ML cs.LG

Leonardo D. González , Victor M. Zavala This is my paper · ORCID
classification stat.MLcs.LG
keywords Bayesianoptimizationgrey-boxmodelingcompositefunctionsGaussianprocessesadaptivelinearizationlowerconfidenceboundchemicalprocessinterconnectedsystems
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 implements and benchmarks BOIS, a Bayesian optimization strategy for composite objective functions of the form $f(x,y(x))$, where the outer function $f$ is known and the intermediate vector $y(x)$ is learned from data with Gaussian process models. Propagating the Gaussian uncertainty in $y$ through a nonlinear $f$ is generally intractable; existing grey-box methods draw many Monte Carlo samples (MC-BO) or solve a larger constrained problem over an augmented space (OP-BO). BOIS instead linearizes $f$ in $y$ around an adaptively updated reference point, which converts the propagation into closed-form algebraic expressions for the mean and standard deviation of $f$. On two chemical-process benchmarks the paper reports that BOIS matches or beats MC-BO and OP-BO in solution quality while spending far less time optimizing its acquisition function, and that its moment estimates agree with high-sample Monte Carlo at a fraction of the cost.

What carries the argument

The load-bearing mechanism is the adaptive linearization, a local Laplace approximation of $f$. At each candidate $x$, the Gaussian process supplies the mean $\hat{y}^\ell$ and covariance $\hat{\Sigma}^\ell$ for $y$; feasibility bounds are enforced by clipping $\hat{y}^\ell$ to its allowable range; a reference point $\hat{y}_0^\ell$ is chosen within an $\epsilon$-neighborhood; and the Jacobian $J=\nabla_y h(x,\hat{y}_0^\ell)$ is evaluated. This single linear map turns the GP's Gaussian density on $y$ into an approximate Gaussian density on $f$, giving the closed-form moments in (19) that feed the lower confidence bound acquisition function. The linearization is rebuilt at every candidate point rather than extrapolated from previous points, and it supports nested GP structures in which some intermediate variables depend on others, which the paper uses to shrink the input spaces of the surrogate models.

What would settle it

On a synthetic composite with deliberately strong nonlinearity in $h$ (for example, $f(x,y)=g(x)+\exp(y_1)+\sin(y_2)$), compute the BOIS moments in (19) and compare them with Monte Carlo estimates using $10^4$ samples at 500 points; if the error in $\sigma_f^\ell$ grows with the GP variance and the acquisition function preferentially samples where the linearization is poor, the central premise fails. A cheaper check is near-boundary behavior: the paper's parity plots already show a visible bias in $\sigma$ when the GP mean is clipped to a feasibility bound, so on a problem where the optimum lies near such a bound, one can test whether that bias changes the selected point.

Watch

Extended reading notes

Core claim

The central claim is that for a once-differentiable composite $f(x,y(x)) = g(x)+h(x,y)$, a local first-order Taylor expansion $f \approx g(x)+h(x,\hat{y}_0^\ell)+J^T(y(x)-\hat{y}_0^\ell)$, with $J=\nabla_y h$, is enough to drive grey-box Bayesian optimization. Because Gaussian densities are closed under affine maps, the framework obtains closed-form moments $m_f^\ell(x) = J^T \hat{y}^\ell + g(x) + h(x,\hat{y}_0^\ell) - J^T \hat{y}_0^\ell$ and $\sigma_f^\ell(x) = (J^T \hat{\Sigma}^\ell J)^{1/2}$, where $\hat{y}^\ell$ is the GP mean clipped to feasibility bounds and $\hat{\Sigma}^\ell$ is the GP covariance. The paper argues that this makes the acquisition function cheaper to evaluate than Monte Carlo-based LCB-CF, and cheaper to optimize than the OP-BO auxiliary problem because optimization stays in the original $x$-space rather than an augmented $(x,y)$-space. On its two benchmarks this yields solutions as good as or better than MC-BO and OP-BO, with acquisition optimization about 41% faster than MC-BO and about 3.3 times faster than OP-BO in the first case study.

Load-bearing premise

The method stands on the assumption that a first-order Taylor expansion of the known function $f(x,y)$ around a single reference value of $y$ stays accurate across the whole region where the Gaussian process is uncertain, which the paper itself notes degrades far from the reference point and near the feasibility bounds of $y$.

Editorial extensions

If this is right

  • If the closed-form moments are accurate, composite-function BO can exploit physics and connectivity without Monte Carlo, making grey-box optimization practical when each system query is expensive.
  • Because the acquisition function is optimized in the original $x$-space rather than an augmented $(x,y)$-space, BOIS remains cheap as the number of intermediate variables grows, and the paper shows this advantage widening in the higher-dimensional of its two case studies.
  • Feasibility constraints on intermediates are handled directly through the clipping step (17), so infeasible regions of $y$ do not need to be learned by sampling.
  • The same propagation step can feed other acquisition functions built from the closed-form moments, such as probability-of-improvement or expected-improvement variants, without changing the linearization machinery.

Reading between the lines

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

  • The paper's parity experiments suggest that BOIS's advantage over MC-BO is not only speed: with $S=1000$ Monte Carlo samples, MC-BO matches BOIS's accuracy but at roughly two orders of magnitude more compute, so the practical gain transfers to any setting where acquisition evaluation dominates the budget.
  • For strongly nonlinear $h$, such as exponential or trigonometric terms, the first-order expansion may understate the true variance in regions of high GP uncertainty; an immediate test would compare BOIS's $\sigma$ to high-precision Monte Carlo on synthetic composites with known curvature.
  • The near-boundary bias the authors observe motivates a natural extension: replace the symmetric GP with a warped or otherwise non-Gaussian surrogate, or use the analytic moments only away from feasibility bounds while falling back to Monte Carlo near them.
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 / 6 minor

Summary. The paper presents BOIS, a grey-box Bayesian optimization method for composite objective functions of the form f(x, y(x)), where y is an unknown intermediate vector-valued function modeled by Gaussian processes. The core contribution is a closed-form approximation of the mean and variance of f obtained by a first-order Taylor expansion of f in y around a local reference point, which yields an LCB-style acquisition function whose evaluation avoids Monte Carlo sampling or the augmented-space optimization used by OP-BO. The manuscript extends the authors' earlier BOIS preprint with feasibility handling for intermediate variables and nested GP structures, and benchmarks the method against standard BO, MC-BO, and OP-BO on two chemical process case studies: a reactor/separator/recycle process and a photobioreactor design problem. The reported results indicate that BOIS matches or improves on the solution value of the comparators while requiring less time for acquisition-function optimization, and the authors provide parity plots showing that BOIS moment estimates agree with high-sample MC-BO estimates.

Significance. If the empirical claims are substantiated, the paper makes a useful contribution by offering a computationally efficient way to propagate uncertainty through nonlinear composite functions in BO, which is relevant for exploiting process knowledge in engineering applications. The derivation of the closed-form moments in Eq. (19) is straightforward and clearly presented, and the authors provide reproducible code and data on GitHub, which is a genuine strength. The two case studies are nontrivial and realistic, and the authors are transparent about the known inaccuracy of the Laplace approximation near feasibility bounds. However, the significance is currently tempered by three issues: the MC-BO baseline is run with an admitted underpowered sample size; the performance differences are reported without statistical significance measures; and the moment-accuracy validation is an internal consistency check against MC-BO on the same GP posterior rather than against an external ground truth. These issues leave the central claim 'as well as or better than existing grey-box methods' only partially supported, although the computational advantage over OP-BO is more plausible.

major comments (4)
  1. [Section 5.1 and Section 5.2] The MC-BO baseline is run with S=100 samples, and the text in Section 5.1 explicitly concedes that 'at a given point x with S=100, different evaluations of the LCB-MCBO AF could return values that differed by over 10%.' This is an admission that the comparator is underpowered, so the reported average improvements over MC-BO (1.2% and 3.3% in Study 1, 1.6% in Study 2) do not establish that BOIS outperforms a properly calibrated MC-BO. Furthermore, these differences are reported as point averages over 25 and 125 trials without error bars, confidence intervals, or paired significance tests, so sampling noise cannot be ruled out. The authors should rerun the MC-BO comparison with a converged sample size (or an adaptive budget) and report the performance differences with appropriate uncertainty quantification, for example paired bootstrap or Wilcoxon tests.
  2. [Section 5.1, Figure 9 and Section 5.2, Figure 12] The moment-accuracy validation compares BOIS's closed-form estimates of m_f(x) and sigma_f(x) against MC-BO estimates computed on the same trained GP posterior of y(x), rather than against true sampled cost values or an independent ground truth. Since the MC-BO reference is itself an estimator of the GP posterior moments (and is noisy at the S=100 baseline), the parity plots demonstrate internal consistency only; they do not validate that the linearization captures the true distribution of the cost function. The authors should validate the moments against direct samples of f(x, y(x)) (e.g., by sampling y from a high-fidelity simulator or the true system) or clearly state that only internal consistency is being checked.
  3. [Section 4, Eqs. (17)-(19)] The closed-form variance in Eq. (19b) uses the GP posterior covariance matrix Σ^ℓ of y, but the paper describes a nested structure in which downstream GP models are evaluated using the current means of upstream GPs, without propagating upstream predictive uncertainty into the downstream inputs. As a result, the covariance used in Eq. (19b) may understate the true uncertainty of f(x, y(x)) in the nested setting, which is precisely the setting used in Study 1 to reduce input dimensionality. This is a potential gap in the uncertainty propagation that should be either justified analytically or quantified empirically, for example by comparing the closed-form variance against a fully propagated Monte Carlo estimate in a nested example.
  4. [Section 5.2, near-bound behavior] The manuscript acknowledges that the Laplace approximation is not accurate near the feasibility bounds of the intermediate functions because the GP's symmetric Gaussian density assigns mass to infeasible values. Since the BOIS acquisition function is built from these moments, the statement that 'this was not an issue' is supported only by the aggregate benchmark results, not by any targeted analysis. The authors should either provide a sensitivity study on a problem whose optimum lies near a feasibility bound, or moderate the claim by explicitly stating that the method's accuracy near bounds remains an open limitation that could affect acquisition decisions in constrained problems.
minor comments (6)
  1. [Section 5.2, first paragraph] The placeholder citation '(see ??)' should be replaced with a proper reference to the literature models for biogas production and CB harvesting.
  2. [Section 5.1 and Section 5.2] The phrases 'by 1.2% and 3.3% respectively' and 'by and average 5.4%' contain grammatical or typographical errors; the latter should read 'by an average of 5.4%'.
  3. [Section 4, Eq. (18)] The reference point notation ŷ_0^ℓ is used in Eqs. (18) and (19) but is not explicitly defined before its first use; the earlier expansion in Eq. (15) uses y_0. Please introduce the notation consistently.
  4. [Section 5, paragraph 3] The statement that 'ϵ in BOIS was set to ŷ × 10^-3' is ambiguous for vector-valued ŷ; specify whether the scaling is applied elementwise or uses a norm.
  5. [References] Reference [14] contains a misspelling: 'Baysian Optimization' should be 'Bayesian Optimization'.
  6. [Algorithm 2] The notation m_f,s^ℓ(x) for individual sampled values of f is confusingly close to the notation m_f^ℓ(x) for the estimated mean; consider using f_s(x) for the samples.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the closed-form moment formulas follow from the stated Taylor linearization and are validated against MC-BO estimates on the same GP; self-citation to the BOIS preprint is historical, not load-bearing.

full rationale

The paper's derivation is self-contained. Equations (14)-(19) obtain the closed-form moment estimates by a first-order Taylor expansion of f with respect to y and the Gaussian affine closure; no parameter is fitted to the mean or variance that is later claimed as a prediction. The reference point y-hat-0 is chosen from an epsilon-neighborhood as stated, not tuned to match the target moments. The parity checks in Figures 9 and 12 compare BOIS moment estimates with MC-BO estimates using the same trained GP model, with S=1000 serving as a numerical reference; this is a validation exercise, not a fit, and the paper explicitly shows the estimates converge as S increases. The benchmark claims are evaluated against known optima of two chemical process models, and the speed advantage is supported by operation counts: BOIS uses one GP draw and one f evaluation per point, whereas MC-BO uses S draws and S evaluations, and OP-BO solves a higher-dimensional constrained problem. The only self-citation, reference [15], attributes the origin of the BOIS framework but is not load-bearing in this manuscript, because Section 4 restates the full derivation of the linearization and moment propagation. The limitations the authors acknowledge, namely that MC-BO with S=100 is noisy and that the Laplace approximation loses accuracy near feasibility bounds, are correctness and calibration risks rather than circularity. No equation reduces by construction to an input, and no prediction is a renamed fitted parameter.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central method relies on standard Gaussian linearization, on the assumption that GP predictive distributions are accurate for the intermediates, and on an ad hoc single-point Laplace approximation. The main free parameters are the linearization neighborhood ϵ, the exploration weight κ, and the MC-BO sample count S, none of which are varied in sensitivity analyses. No new physical entities are introduced.

free parameters (3)
  • epsilon (ϵ) = ŷℓ × 10⁻³
    Defines the neighborhood for the linearization reference point ŷ0ℓ. Chosen by hand, with no sensitivity analysis reported.
  • LCB exploration weight κ = not specified
    Used in all acquisition functions and the OP-BO confidence bounds. The value is not reported and no sensitivity study is given, so the comparison could depend on an unstated tuning choice.
  • MC-BO sample count S = 100
    Fixed for the MC-BO baseline without a convergence check. The paper itself notes that S=100 can produce AF values varying by over 10%, which likely disadvantages MC-BO.
assumptions (4)
  • standard math Gaussian closure under linear transformations: if y ~ N(m, Σ), then aᵀ y + b is Gaussian with moments aᵀm + b and aᵀΣa.
    Used to derive the closed-form mean and variance in Eq. (10) and Eq. (19).
  • domain assumption The intermediate functions y(x) are modeled as independent single-output Gaussian processes with Gaussian predictive distributions.
    The GP predictive distribution is treated as the correct description of the unknown intermediates. The paper does not validate this against real system data.
  • ad hoc to paper f is once-differentiable in y and a first-order Taylor expansion around one reference point is sufficient over the support of the GP posterior.
    This is the Laplace approximation at the heart of BOIS. The authors acknowledge it degrades near feasibility bounds and for nonlinear f.
  • ad hoc to paper Nested intermediates can be evaluated by plugging in the mean of upstream GP models, without propagating upstream uncertainty into downstream GP inputs.
    Figure 5 and the sequential evaluation described in Section 4 substitute the current mean of upstream y into downstream models; the resulting Σ̂ℓ may not represent the full joint covariance of y.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On the Implementation of a Bayesian Optimization Framework for Interconnected Systems." pith.science (2026). https://pith.science/paper/WWL2ZQE3

@misc{pith2026250100967,
  author       = {Pith},
  title        = {Pith review of: On the Implementation of a Bayesian Optimization Framework for Interconnected Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WWL2ZQE3}},
  note         = {Machine review of arXiv:2501.00967}
}
abstract

Bayesian optimization (BO) is an effective paradigm for the optimization of expensive-to-sample systems. Standard BO learns the performance of a system $f(x)$ by using a Gaussian Process (GP) model; this treats the system as a black-box and limits its ability to exploit available structural knowledge (e.g., physics and sparse interconnections in a complex system). Grey-box modeling, wherein the performance function is treated as a composition of known and unknown intermediate functions $f(x, y(x))$ (where $y(x)$ is a GP model) offers a solution to this limitation; however, generating an analytical probability density for $f$ from the Gaussian density of $y(x)$ is often an intractable problem (e.g., when $f$ is nonlinear). Previous work has handled this issue by using sampling techniques or by solving an auxiliary problem over an augmented space where the values of $y(x)$ are constrained by confidence intervals derived from the GP models; such solutions are computationally intensive. In this work, we provide a detailed implementation of a recently proposed grey-box BO paradigm, BOIS, that uses adaptive linearizations of $f$ to obtain analytical expressions for the statistical moments of the composite function. We show that the BOIS approach enables the exploitation of structural knowledge, such as that arising in interconnected systems as well as systems that embed multiple GP models and combinations of physics and GP models. We benchmark the effectiveness of BOIS against standard BO and existing grey-box BO algorithms using a pair of case studies focused on chemical process optimization and design. Our results indicate that BOIS performs as well as or better than existing grey-box methods, while also being less computationally intensive.

Figures

Figures reproduced from arXiv: 2501.00967 by the authors.

Figure 1
Figure 1. Grey-box systems often exhibit a known structure where the connectivity between differ [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the adaptive linearization scheme employed by BOIS. At a point [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Workflow of the S-BO framework. Using a dataset [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Workflow of the OP-BO algorithm. The data in [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Schematic representation of a nested function structure for y. Using a first-order Taylor series expansion, we linearize (14) with respect to y around a reference point y0: f(x, y(x)) ≈ g(x) + h(x, y0) + J T (y − y0) (15) where the Jacobian is: J = ∇yh(x, y0) (16a) = ∇…
Figure 6
Figure 6. Figure 6: Workflow of the BOIS algorithm. Here, we note that [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Performance comparison of the tested algorithms for the chemical process optimization [PITH_FULL_IMAGE:figures/full_fig_p018_7.png]
Figure 8
Figure 8. Figure 8: Computational intensity of the tested algorithms for solving the chemical process opti￾mization problem measured as (a) the total execution time and (b) the difference between the total execution time and total system sampling time. We further quantified the variations…
Figure 9
Figure 9. Figure 9: Parity plots of the estimates of mℓ f (x) (a) and σ ℓ f (x) (b) for (21) and log10 of the time required to generate the estimates (c) at 500 points in X using BOIS with ϵ = ˆy × 10−3 and MC-BO with samples sizes S = 10, 102 , and 103 ; the same trained GP model of y(x)…
Figure 10
Figure 10. Figure 10: Performance comparison of the tested algorithms for the photobioreactor design prob [PITH_FULL_IMAGE:figures/full_fig_p023_10.png]
Figure 11
Figure 11. Figure 11: Computational intensity of the tested algorithms for solving the photobioreactor design problem measured as (a) the total execution time and (b) the difference between the total execution time and total system sampling time [PITH_FULL_IMAGE:figures/full_fig_p024_11.png]
Figure 12
Figure 12. Figure 12: Parity plots of the estimates of mℓ f (x) (a) and σ ℓ f (x) (b) for the MSP function and log10 of the time required to generate the estimates (c) at 500 points in X using BOIS with ϵ = ˆy × 10−3 and MC-BO with samples sizes S = 10, 102 , and 103 ; the same trained GP …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 39 canonical work pages

  1. [1]

    M. A. Alvarez, L. Rosasco, and N. D. Lawrence. Kernels for vector-valued functions: a review. arXiv preprint arXiv:1106.6251, 2012

    arXiv 2012

  2. [2]

    Astudillo and P

    R. Astudillo and P . Frazier. Bayesian optimization of composite functions. In K. Chaudhuri and R. Salakhutdinov, editors,Proceedings of the 36th International Conference on Machine Learn- ing, volume 97 of Proceedings of Machine Learning Research , pages 354–363. PMLR, 09–15 Jun 2019

    work page 2019

  3. [3]

    Astudillo and P

    R. Astudillo and P . Frazier. Bayesian optimization of function networks. Advances in neural information processing systems, 34:14463–14475, 2021

    work page 2021

  4. [4]

    Astudillo and P

    R. Astudillo and P . Frazier. Thinking inside the box: A tutorial on grey-box Bayesian opti- mization. In Proceedings of the 2021 Winter Simulation Conference, December 2021. 28 http://zavalab.engr.wisc.edu

    work page 2021

  5. [5]

    Bajaj, S

    I. Bajaj, S. S. Iyer, and M. F. Hasan. A trust region-based two phase algorithm for constrained black-box and grey-box optimization with infeasible initial point. Computers & Chemical En- gineering, 116:306–321, 2018

    work page 2018

  6. [6]

    Balandat, B

    M. Balandat, B. Karrer, D. Jiang, S. Daulton, B. Letham, A. Wislon, and E. Bashky. BOTORCH: A framework for efficient Monte-Carlo Bayesian optimization. In Proceedings of the 34th Con- ference International Conference on Neural Information Processing Systems, NIPS ’20, pages 21524– 21538. Curran Associates Inc., Dec 2020

    work page 2020

  7. [7]

    Baur and V

    W. Baur and V . Strassen. The complexity of partial derivatives. Theoretical Computer Science, 22(3):317–330, 1983

    work page 1983

  8. [8]

    Beykal, F

    B. Beykal, F. Boukouvala, C. A. Floudas, and E. N. Pistikopoulos. Optimal design of en- ergy systems using constrained grey-box multi-objective optimization. Computers & Chemical Engineering, 116:488–502, 2018

    work page 2018

Show all 43 references
  1. [9]

    Boukouvala and C

    F. Boukouvala and C. Floudas. ARGONAUT: AlgoRithms for Global Optimization of coN- strAined grey-box compUTational problems. Optimization Letters, 11(5):895–913, 2017

    2017

  2. [10]

    Brochu, V

    E. Brochu, V . M. Cora, and N. De Freitas. A tutorial on Bayesian optimization of expen- sive cost functions, with application to active user modeling and hierarchical reinforcement learning. arXiv preprint arXiv:1012.2599, 2010

    2010 arXiv

  3. [11]

    Chakrabarty, S

    A. Chakrabarty, S. A. Bortoff, and C. R. Laughman. Simulation failure-robust Bayesian op- timization for data-driven parameter estimation. IEEE Transactions on Systems, Man, and Cy- bernetics: Systems, 53(5):2629–2640, 2023

    2023

  4. [12]

    R. L. Clark, L. L. McGinley, H. M. Purdy, T. C. Korosh, J. L. Reed, T. W. Root, and B. F. Pfleger. Light-optimized growth of cyanobacterial cultures: growth phases and productiv- ity of biomass and secreted molecules in light-limited batch growth. Metabolic engineering, 47:230...

    2018

  5. [13]

    A. Conn, K. Scheinberg, and L. Vicente. Introduction to Derivative-free Optimization . SIAM, 2009

    2009

  6. [14]

    R. Garnett. Baysian Optimization. Cambridge University Press, 2023

    2023

  7. [15]

    L. D. Gonz ´alez and V . M. Zavala. BOIS: Bayesian Optimization of Interconnected Systems. arXiv preprint arXiv:2311.11254, 2023

    2023 arXiv

  8. [16]

    Greenhill, S

    S. Greenhill, S. Rana, S. Gupta, P . Vellanki, and S. Venkatesh. Bayesian optimization for adaptive experimental design: A review. IEEE access, 8:13937–13948, 2020

    2020

  9. [17]

    Griewank and A

    A. Griewank and A. Walther. Evaluating Derivatives: Principles and Techniques of Algorithmic Differentiation. SIAM, Philadelphia, 2008. 29 http://zavalab.engr.wisc.edu

    2008

  10. [18]

    F. Hase, M. Aldeghi, R. J. Hickman, L. M. Roch, and A. Aspuru-Guzik. Gryffin: An algorithm for Bayesian optimization of categorical variables informed by expert knowledge. Applied Physics Reviews, 8(3):031406, 2021

    2021

  11. [19]

    D. R. Jones, M. Schonlau, and W. J. Welch. Efficient global optimization of expensive black- box functions. Journal of Global Optimization, 13(4):455–492, 1998

    1998

  12. [20]

    Kandasamy, G

    K. Kandasamy, G. Dasarathy, J. Schnieder, and B. P ´ozcos. Multi-fidelity Bayesian optimisa- tion with continuous approximations. In D. Precup and Y. Teh, editors,Uncertainty in Artificial Intelligence, volume 70 of Proceedings of Machine Learning Research , pages 1799–1808. P...

    2017

  13. [21]

    R. R. Lam, M. Poloczek, P . I. Frazier, and K. E. Willcox. Advances in Bayesian optimiza- tion with applications in aerospace engineering. In Proceedings of the AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, Boston, MA, 2018. AIAA

    2018

  14. [22]

    H. Liu, J. Cai, and Y.-S. Ong. Remarks on multi-output Gaussian process regression. Knowedge-Based Systemsl, 144:102–121, 2018

    2018

  15. [23]

    Lu and J

    C. Lu and J. A. Paulson. No-regret constrained Bayesian optimization of noisy and expen- sive hybrid models using differentiable quantile function approximations. Journal of Process Control, 131:103085, 2023

    2023

  16. [24]

    Q. Lu, L. Gonz ´alez, R. Kumar, and V . Zavala. Bayesian optimization with reference models: A case study in MPC for HVAC central plants.Computers & Chemical Engineering, 154:107491, 2021

    2021

  17. [25]

    Mat ´ern

    B. Mat ´ern. Spatial variation : Stochastic models and their application to some problems in forest surveys and other sampling investigations. In Messages from the State Forestry Research Institute, volume 49, 1960

    1960

  18. [26]

    J. Mockus. Bayesian Approach to Global Optimization: Theory and Applications. Springer Science & Business Media, 2012

    2012

  19. [27]

    Paulson and C

    J. Paulson and C. Lu. COBALT: COnstrained Bayesian optimizAtion of computaionaLly ex- pensive grey-box models exploiting derivaTive information. Computers & Chemical Engineer- ing, 160:107700, 2022

    2022

  20. [28]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V . Michel, B. Thirion, O. Grisel, M. Blondel, P . Pret- tenhofer, R. Weiss, V . Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Per- rot, and E. Duchesnay. Scikit-learn: Machine learning in Python. Journal of Machine L...

    2011

  21. [29]

    Priem, N

    R. Priem, N. Bartoli, and Y. Diouane. On the use of upper trust bounds in constrained Bayesian optimization infill criterion. In AIAA Aviation 2019 Forum , pages 1–10, Dallas, United States, June 2019. 30 http://zavalab.engr.wisc.edu

    2019

  22. [30]

    Radivojevi ´c, Z

    T. Radivojevi ´c, Z. Costello, K. Workman, and H. Garcia Martin. A machine learning auto- mated recommendation tool for synthetic biology. Nature Communications, 11(1):4879, 2020

    2020

  23. [31]

    C. E. Rasmussen and C. K. I. Williams. Gaussian Processes for Machine Learning . MIT Press, 2006

    2006

  24. [32]

    Shahriari, K

    B. Shahriari, K. Swersky, Z. Wang, R. Adams, and N. de Freitas. Taking the human out of the loop: A review of Bayesian optimization. Proceedings of the IEEE, 104:148–175, 2016

    2016

  25. [33]

    Snelson, Z

    E. Snelson, Z. Ghahramani, and C. Rasmussen. Warped Gaussian processes. In S. Thrun, L. Saul, and B. Sch ¨olkopf, editors, Advances in Neural Information Processing Systems , vol- ume 16, pages 337–334. MIT Press, 2004

    2004

  26. [34]

    Snoek, H

    J. Snoek, H. Larochelle, and R. P . Adams. Practical Bayesian optimization of machine learning algorithms. In Advances in Neural Information Processing Systems, 2012

    2012

  27. [35]

    Sohlberg and E

    B. Sohlberg and E. Jacobsen. Grey-box modeling – branches and experiences.IFAC Proceedings Volumes, 41(2):11415–11420, 2008. 17th IFAC World Congress

    2008

  28. [36]

    Sorourifar, N

    F. Sorourifar, N. Choksi, and J. A. Paulson. Computationally efficient integrated design and predictive control of flexible energy systems using multi-fidelity simulation-based Bayesian optimization. Journal of Optimal Control Applications and Methods, 44(4):549–576, 2023

    2023

  29. [37]

    Sorourifar, G

    F. Sorourifar, G. Makrygirgos, A. Mesbah, and J. A. Paulson. A data-driven automatic tun- ing method for MPC under uncertainty using constrained Bayesian optimization. IFAC- PapersOnLine, 54(3):243–250, 2021. 16th IFAC Symposium on Advanced Control of Chemical Processes ADCHEM 2021

    2021

  30. [38]

    Thompson, V

    J. Thompson, V . Zavala, and O. Venturelli. Integrating a tailored recurrent neural network with Bayesian experimental design to optimize microbial community functions. PLOS Com- putational Biology, 19(9):1–25, 2023

    2023

  31. [39]

    Uhrenholt and B

    A. Uhrenholt and B. Jensen. Efficient Bayesian optimization for target vector estimation. In K. Chaudhuri and M. Sugiyama, editors, Proceedings of the Twenty-Second International Con- ference on Artificial Intelligence and Statistics , volume 89 of Proceedings of Machine Learn...

    2019

  32. [40]

    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

  33. [41]

    Wilson, A

    A. Wilson, A. Fern, and P . Tadepalli. Using trajectory data to improve Bayesian optimization for reinforcement learning. Journal of Machine Learning Research, 15(8):253–282, 2014

    2014

  34. [42]

    J. Wu, S. Toscano-Palmerin, P . I. Frazier, and A. G. Wilson. Practical multi-fidelity Bayesian optimization for hyperparameter tuning. arXiv preprint arXiv:1903.04703, 2019

    1903 arXiv

  35. [43]

    W. Xu, Y. Jiang, B. Svetozarevic, and C. N. Jones. Bayesian optimization of expensive nested grey-box functions. arXiv preprint arXiv:2306.05150, 2023. 32

    2023 arXiv

Pith tools

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