pith. sign in

arxiv: 2606.19230 · v1 · pith:4XM7X7OWnew · submitted 2026-06-17 · 💻 cs.LG · cs.HC· stat.ML

A Human-in-the-Loop Bayesian Optimization Framework for Constraint-Aware Bioprocess Development

Samuel Stricker , Claus Wirnsperger , Alessandro Butt\'e , Laura Helleckes , Gonzalo Guill\'en Gos\'albez , Antonio del Rio Chanona , Mehmet Mercang\"oz This is my paper

Pith reviewed 2026-06-26 21:10 UTC · model grok-4.3

classification 💻 cs.LG cs.HCstat.ML
keywords Bayesian optimizationhuman-in-the-loopPareto frontconstraint satisfactionrobust optimizationGaussian processesbioprocess developmentCHO cell culture
0
0 comments X

The pith

Extending Pareto-guided sampling lets experts interactively balance performance, uncertainty, constraints and robustness in Bayesian bioprocess optimization

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

The paper extends Pareto Front Guided Sampling by treating the posterior probability of satisfying output constraints and Monte Carlo estimates of performance under input perturbations as additional objectives alongside predicted performance and model uncertainty. These quantities are computed from a Gaussian process surrogate and the resulting multi-objective Pareto front is shown to a domain expert through pairwise two-dimensional projections on an interactive dashboard. The approach is demonstrated on an eight-dimensional fed-batch CHO cell culture simulator, where it supports systematic selection of operating conditions that are high-performing, constraint-compliant and resilient to implementation variability while allowing stopping criteria to evolve with the model.

Core claim

The framework reformulates Gaussian process surrogate-derived quantities as objectives of a multi-objective optimization problem; the posterior probability of satisfying output specification limits is computed analytically from the GP posterior and incorporated as an explicit Pareto objective, while a Monte Carlo strategy estimates expected lower-confidence performance over user-defined input perturbations; the resulting multi-dimensional Pareto representation is exposed to a domain expert for interactive candidate selection rather than returning a single automated recommendation.

What carries the argument

Multi-dimensional Pareto front whose objectives are GP-predicted performance, model uncertainty, analytically computed probability of constraint satisfaction, and Monte Carlo estimate of robustness to input perturbations

If this is right

  • Pairwise two-dimensional projections make simultaneous trade-offs between performance, uncertainty, feasibility and robustness visible to the expert
  • Expert selection criteria can be iteratively refined as the surrogate improves and development objectives evolve
  • The dashboard provides a principled, expert-defined stopping criterion for experimental resource allocation
  • High-performing, feasibility-compliant and perturbation-resilient operating conditions can be identified systematically in the eight-dimensional fed-batch example

Where Pith is reading between the lines

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

  • The interactive dashboard could be combined with sequential experimental data to update the surrogate and re-compute the Pareto front after each new measurement
  • Similar reformulation of surrogate quantities into Pareto objectives might apply to other surrogate-based optimization settings that require explicit handling of safety constraints and implementation uncertainty
  • The Monte Carlo robustness estimate could be replaced by an analytic approximation if the input perturbation distribution is Gaussian, potentially reducing computational cost

Load-bearing premise

The Gaussian process surrogate model provides sufficiently accurate posterior distributions to allow reliable analytical computation of constraint satisfaction probabilities and meaningful Monte Carlo estimates of robustness under input perturbations.

What would settle it

Running the conditions selected from the dashboard on the actual CHO simulator or real process and observing that the measured constraint satisfaction rates or robustness levels fall outside the ranges predicted by the Pareto front would falsify the framework's utility for informed selection.

Figures

Figures reproduced from arXiv: 2606.19230 by Alessandro Butt\'e, Antonio del Rio Chanona, Claus Wirnsperger, Gonzalo Guill\'en Gos\'albez, Laura Helleckes, Mehmet Mercang\"oz, Samuel Stricker.

Figure 1
Figure 1. Figure 1: Representative classical experimental design strategies. (a) Factorial designs sample combinations of factor [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Basic PFGS workflow for titer maximization: pink points indicate performed experiments, blue points [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Probabilistic calculation of CQA constraint satisfaction for a candidate recipe. The GP predictive distribution [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Restricted constrained Pareto front for the first PFGS iteration with a purity requirement of [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Robustness strategies used to extend PFGS beyond nominal recipe optimization. (a) Worst-case robustness [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Example of robustness-based candidate interpretation. (left) Pareto representation of nominal predicted titer [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Schematic of the bioreactor process, including the initial setup, two bolus feeds, and the final harvest. [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Two-dimensional slice of the bioprocess simulator with all non-displayed inputs fixed at their default values. [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Evolution of the Pareto front throughout the optimization campaign. Blue points denote Pareto-optimal [PITH_FULL_IMAGE:figures/full_fig_p016_9.png] view at source ↗
read the original abstract

This work presents an extension to Pareto Front Guided Sampling (PFGS), a Human-in-the-Loop (HitL) Bayesian Optimization (BO) framework in which Gaussian process (GP) surrogate-derived quantities are reformulated as objectives of a multi-objective optimization problem, and the resulting Pareto front is exposed to a domain expert for interactive candidate selection rather than returning a single automated recommendation. The framework is extended in two directions: constrained optimization is addressed by incorporating the posterior probability of satisfying output specification limits as an explicit Pareto objective, computed analytically from the GP posterior distribution; robust optimization is addressed by a Monte Carlo sampling strategy that estimates expected lower-confidence performance over a user-defined variability of input perturbations, capturing performance degradation under likely implementation deviations. The resulting multi-dimensional Pareto representation renders trade-offs between predicted performance, model uncertainty, probabilistic constraint satisfaction, and input robustness simultaneously visible through pairwise two-dimensional projections on an interactive dashboard, enabling selection criteria to be iteratively refined as the surrogate model improves and development objectives evolve. The framework is showcased on an eight-dimensional fed-batch Chinese Hamster Ovary (CHO) cell culture simulator demonstrating systematic identification of high-performing, feasibility-compliant, and perturbation-resilient operating conditions, and illustrating how expert-defined requirements provide a principled stopping criterion and support informed allocation of experimental resources.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The manuscript extends the Pareto Front Guided Sampling (PFGS) human-in-the-loop Bayesian optimization framework. Gaussian process surrogate quantities are reformulated as objectives in a multi-objective optimization problem whose Pareto front is exposed to a domain expert via an interactive dashboard for candidate selection. Two extensions are introduced: constrained optimization via the analytically computed posterior probability of satisfying output limits, and robust optimization via Monte Carlo estimation of expected lower-confidence performance under input perturbations. The resulting multi-dimensional Pareto representation is demonstrated on an eight-dimensional fed-batch CHO cell culture simulator, where it supports systematic identification of high-performing, feasibility-compliant, and perturbation-resilient conditions together with expert-defined stopping criteria.

Significance. If the implementation details hold, the framework supplies a usable interface for incorporating probabilistic constraints and input robustness directly into the selection process of Bayesian optimization, which is relevant for bioprocess applications. The interactive dashboard that renders pairwise projections of the multi-objective trade-offs is a concrete contribution that allows iterative refinement of selection criteria. The simulator demonstration is a legitimate vehicle for illustrating the workflow; the approach relies on standard GP posterior properties and does not exhibit internal circularity or unsupported derivations.

minor comments (3)
  1. [Methods] The manuscript should add a short paragraph in the methods section clarifying the precise definition of the Monte Carlo robustness objective (number of samples, perturbation distribution, and how the lower-confidence bound is aggregated) to allow exact reproduction.
  2. [Results] Figure captions for the dashboard projections should explicitly state the number of objectives being visualized and the mapping from the four-dimensional Pareto set to the displayed pairwise planes.
  3. [Discussion] A brief discussion of computational cost for the analytical constraint probability versus the Monte Carlo robustness estimate would help readers assess scalability beyond the eight-dimensional simulator.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the detailed and positive assessment of our manuscript, including the accurate summary of the PFGS extensions for probabilistic constraints and input robustness. The recommendation for minor revision is noted; we will prepare a revised version addressing any editorial or minor points that may arise during production.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The described framework applies standard GP posterior properties to compute analytical constraint satisfaction probabilities and uses Monte Carlo sampling for robustness estimates under input perturbations. These quantities are then treated as additional objectives in a multi-objective optimization whose Pareto front is presented to a human expert. No derivation reduces a prediction to a fitted input by construction, no self-citation is invoked as a uniqueness theorem, and no ansatz is smuggled via prior work. The central claim is an engineering extension of existing BO techniques rather than a closed mathematical reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no concrete information on free parameters, axioms, or invented entities; all fields left empty.

pith-pipeline@v0.9.1-grok · 5796 in / 1123 out tokens · 23665 ms · 2026-06-26T21:10:19.127789+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

45 extracted references · 31 canonical work pages · 2 internal anchors

  1. [1]

    A Guide to Bayesian Optimization in Bioprocess Engineering

    Maximilian Siska et al. “A Guide to Bayesian Optimization in Bioprocess Engineering”. en. In:Biotechnology and Bioengineering123.4 (2026), pp. 805–830.ISSN: 1097-0290.DOI:10.1002/bit.70129

    work page doi:10.1002/bit.70129 2026

  2. [2]

    Bayesian reaction optimization as a tool for chemical synthesis

    Benjamin J. Shields et al. “Bayesian reaction optimization as a tool for chemical synthesis”. en. In:Nature 590.7844 (Feb. 2021), pp. 89–96.ISSN: 1476-4687.DOI:10.1038/s41586-021-03213-y

    work page doi:10.1038/s41586-021-03213-y 2021

  3. [3]

    InProceedings of the 15th ACM SIGPLAN-SIGACT symposium on Principles of programming languages

    Peter I. Frazier and Jialei Wang.Bayesian optimization for materials design. 2016.DOI: 10.1007/978-3-319- 23871-5_3

    work page doi:10.1007/978-3-319- 2016

  4. [4]

    A stochastic data-driven Bayesian optimization approach for intensified ethanol–water separation systems

    Osney Pérez-Ones and Antonio Flores-Tlacuahuac. “A stochastic data-driven Bayesian optimization approach for intensified ethanol–water separation systems”. In:Chemical Engineering and Processing - Process Intensification 197 (Mar. 2024), p. 109708.ISSN: 0255-2701.DOI:10.1016/j.cep.2024.109708

    work page doi:10.1016/j.cep.2024.109708 2024

  5. [5]

    Design of experiments and design space approaches in the pharmaceutical bioprocess optimization

    Alice Kasemiire et al. “Design of experiments and design space approaches in the pharmaceutical bioprocess optimization”. In:European Journal of Pharmaceutics and Biopharmaceutics166 (Sept. 2021), pp. 144–154. ISSN: 0939-6411.DOI:10.1016/j.ejpb.2021.06.004

    work page doi:10.1016/j.ejpb.2021.06.004 2021

  6. [6]

    Bayesian Optimization for Adaptive Experimental Design: A Review

    Stewart Greenhill et al. “Bayesian Optimization for Adaptive Experimental Design: A Review”. In:IEEE Access 8 (2020), pp. 13937–13948.ISSN: 2169-3536.DOI:10.1109/ACCESS.2020.2966228. 18 APREPRINT- JUNE18, 2026

    work page doi:10.1109/access.2020.2966228 2020

  7. [7]

    A Tutorial on Bayesian Optimization

    Peter I. Frazier.A Tutorial on Bayesian Optimization. July 2018.DOI:10.48550/arXiv.1807.02811

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1807.02811 2018

  8. [8]

    Aspects and Implementation of Pharmaceutical Quality by Design from Conceptual Frame- works to Industrial Applications

    Shiwei Yang et al. “Aspects and Implementation of Pharmaceutical Quality by Design from Conceptual Frame- works to Industrial Applications”. en. In:Pharmaceutics17.5 (May 2025), p. 623.ISSN: 1999-4923.DOI: 10.3390/pharmaceutics17050623

    work page doi:10.3390/pharmaceutics17050623 2025

  9. [9]

    Maximizing neotissue growth kinetics in a perfusion bioreactor: An in silico strategy using model reduction and Bayesian optimization

    Mohammad Mehrian et al. “Maximizing neotissue growth kinetics in a perfusion bioreactor: An in silico strategy using model reduction and Bayesian optimization”. en. In:Biotechnology and Bioengineering115.3 (2018), pp. 617–629.ISSN: 1097-0290.DOI:10.1002/bit.26500

    work page doi:10.1002/bit.26500 2018

  10. [10]

    Design of Biopharmaceutical Formulations Accelerated by Machine Learning

    Harini Narayanan et al. “Design of Biopharmaceutical Formulations Accelerated by Machine Learning”. eng. In:Molecular Pharmaceutics18.10 (Oct. 2021), pp. 3843–3853.ISSN: 1543-8392.DOI: 10 . 1021 / acs . molpharmaceut.1c00469

    2021

  11. [11]

    Multi-objective Bayesian algorithm automatically discovers low-cost high-growth serum- free media for cellular agriculture application

    Zachary Cosenza et al. “Multi-objective Bayesian algorithm automatically discovers low-cost high-growth serum- free media for cellular agriculture application”. en. In:Engineering in Life Sciences23.8 (2023), e2300005.ISSN: 1618-2863.DOI:10.1002/elsc.202300005

    work page doi:10.1002/elsc.202300005 2023

  12. [12]

    High-throughput screening of catalytically active inclusion bodies using laboratory automation and Bayesian optimization

    Laura Marie Helleckes et al. “High-throughput screening of catalytically active inclusion bodies using laboratory automation and Bayesian optimization”. en. In:Microbial Cell Factories23.1 (Feb. 2024), p. 67.ISSN: 1475-2859. DOI:10.1186/s12934-024-02319-y

    work page doi:10.1186/s12934-024-02319-y 2024

  13. [13]

    Adrian Martens et al.Holistic Bioprocess Development Across Scales Using Multi-Fidelity Batch Bayesian Optimization. Aug. 2025.DOI:10.48550/arXiv.2508.10970

    work page doi:10.48550/arxiv.2508.10970 2025

  14. [14]

    Bayesian Optimization with Inequality Constraints

    Jacob Gardner et al. “Bayesian Optimization with Inequality Constraints”. en. In:Proceedings of the 31st International Conference on Machine Learning. PMLR, June 2014, pp. 937–945

    2014

  15. [15]

    Bayesian optimization with known experimental and design constraints for chemistry applications

    Riley J. Hickman et al. “Bayesian optimization with known experimental and design constraints for chemistry applications”. en. In:Digital Discovery1.5 (Oct. 2022), pp. 732–744.ISSN: 2635-098X.DOI: 10 . 1039 / D2DD00028H

    2022

  16. [16]

    COBALT: COnstrained Bayesian optimizAtion of computationaLly expensive grey-box models exploiting derivaTive information

    Joel A. Paulson and Congwen Lu. “COBALT: COnstrained Bayesian optimizAtion of computationaLly expensive grey-box models exploiting derivaTive information”. In:Computers & Chemical Engineering160 (Apr. 2022), p. 107700.ISSN: 0098-1354.DOI:10.1016/j.compchemeng.2022.107700

    work page doi:10.1016/j.compchemeng.2022.107700 2022

  17. [17]

    Constrained robust Bayesian optimization of expensive noisy black-box functions with guaranteed regret bounds

    Akshay Kudva, Farshud Sorourifar, and Joel A. Paulson. “Constrained robust Bayesian optimization of expensive noisy black-box functions with guaranteed regret bounds”. en. In:AIChE Journal68.12 (2022), e17857.ISSN: 1547-5905.DOI:10.1002/aic.17857

    work page doi:10.1002/aic.17857 2022

  18. [18]

    Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics

    Felix Berkenkamp, Andreas Krause, and Angela P. Schoellig. “Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics”. en. In:Machine Learning112.10 (Oct. 2023), pp. 3713–3747. ISSN: 0885-6125, 1573-0565.DOI:10.1007/s10994-021-06019-1

    work page doi:10.1007/s10994-021-06019-1 2023

  19. [19]

    Doyle.Safe Bayesian Optimization using Interior-Point Methods – Applied to Personalized Insulin Dose Guidance

    Dinesh Krishnamoorthy and Francis J. Doyle.Safe Bayesian Optimization using Interior-Point Methods – Applied to Personalized Insulin Dose Guidance. May 2022.DOI:10.48550/arXiv.2205.02327

    work page doi:10.48550/arxiv.2205.02327 2022

  20. [20]

    Neural network training as an optimal control problem : — an augmented lagrangian approach —

    Panagiotis Petsagkourakis, Benoit Chachuat, and Ehecatl Antonio del Rio-Chanona. “Safe Real-Time Optimiza- tion using Multi-Fidelity Gaussian Processes”. In:2021 60th IEEE Conference on Decision and Control (CDC). Dec. 2021, pp. 6734–6741.DOI:10.1109/CDC45484.2021.9683599

    work page doi:10.1109/cdc45484.2021.9683599 2021

  21. [21]

    Efficient Tuning of an Isotope Separation Online System Through Safe Bayesian Optimization with Simulation-Informed Gaussian Process for the Constraints

    Santiago Ramos Garces et al. “Efficient Tuning of an Isotope Separation Online System Through Safe Bayesian Optimization with Simulation-Informed Gaussian Process for the Constraints”. en. In:Mathematics12.23 (Jan. 2024), p. 3696.ISSN: 2227-7390.DOI:10.3390/math12233696

    work page doi:10.3390/math12233696 2024

  22. [22]

    Bayesian Optimization Searching for Robust Solutions

    Hoai Phuong Le and Juergen Branke. “Bayesian Optimization Searching for Robust Solutions”. In:2020 Winter Simulation Conference (WSC). Dec. 2020, pp. 2844–2855.DOI:10.1109/WSC48552.2020.9383978

    work page doi:10.1109/wsc48552.2020.9383978 2020

  23. [23]

    Adversarially Robust Optimization with Gaussian Processes

    Ilija Bogunovic et al. “Adversarially Robust Optimization with Gaussian Processes”. In:Advances in Neural Information Processing Systems. V ol. 31. Curran Associates, Inc., 2018

    2018

  24. [24]

    May 2024.DOI: 10

    Dorina Weichert et al.Robust Entropy Search for Safe Efficient Bayesian Optimization. May 2024.DOI: 10. 48550/arXiv.2405.19059

    arXiv 2024

  25. [25]

    Adversarially robust Bayesian optimization for efficient auto-tuning of generic control structures under uncertainty

    Joel A. Paulson, Georgios Makrygiorgos, and Ali Mesbah. “Adversarially robust Bayesian optimization for efficient auto-tuning of generic control structures under uncertainty”. en. In:AIChE Journal68.6 (2022), e17591. ISSN: 1547-5905.DOI:10.1002/aic.17591

    work page doi:10.1002/aic.17591 2022

  26. [26]

    Robust Bayesian optimization for flexibility analysis of expensive simulation-based models with rigorous uncertainty bounds

    Akshay Kudva, Wei-Ting Tang, and Joel A. Paulson. “Robust Bayesian optimization for flexibility analysis of expensive simulation-based models with rigorous uncertainty bounds”. In:Computers & Chemical Engineering 181 (Feb. 2024), p. 108515.ISSN: 0098-1354.DOI:10.1016/j.compchemeng.2023.108515

    work page doi:10.1016/j.compchemeng.2023.108515 2024

  27. [27]

    BONSAI: Structure-exploiting robust Bayesian optimization for networked black-box systems under uncertainty

    Akshay Kudva and Joel A. Paulson. “BONSAI: Structure-exploiting robust Bayesian optimization for networked black-box systems under uncertainty”. In:Computers & Chemical Engineering(Sept. 2025), p. 109393.ISSN: 0098-1354.DOI:10.1016/j.compchemeng.2025.109393. 19 APREPRINT- JUNE18, 2026

    work page doi:10.1016/j.compchemeng.2025.109393 2025

  28. [28]

    Qi Zhang, Kaiyi Ji, and Shaofeng Zou.Distributionally Robust Bayesian Optimization: From Single to Multiple Objectives. en. Oct. 2025

    2025

  29. [29]

    Human-in-the-Loop AI Use in Ongoing Process Verification in the Pharmaceutical Industry

    Miquel Romero-Obon et al. “Human-in-the-Loop AI Use in Ongoing Process Verification in the Pharmaceutical Industry”. en. In:Information16.12 (Dec. 2025), p. 1082.ISSN: 2078-2489.DOI:10.3390/info16121082

    work page doi:10.3390/info16121082 2025

  30. [30]

    Human-algorithm collaborative Bayesian optimization for engineering systems

    Tom Savage and Ehecatl Antonio del Rio Chanona. “Human-algorithm collaborative Bayesian optimization for engineering systems”. In:Computers & Chemical Engineering189 (Oct. 2024), p. 108810.ISSN: 0098-1354. DOI:10.1016/j.compchemeng.2024.108810

    work page doi:10.1016/j.compchemeng.2024.108810 2024

  31. [31]

    Masaki Adachi et al.Looping in the Human: Collaborative and Explainable Bayesian Optimization. en. Feb. 2024

    2024

  32. [32]

    Human–machine collaboration for improving semiconductor process development

    Keren J. Kanarik et al. “Human–machine collaboration for improving semiconductor process development”. en. In:Nature616.7958 (Apr. 2023), pp. 707–711.ISSN: 1476-4687.DOI:10.1038/s41586-023-05773-7

    work page doi:10.1038/s41586-023-05773-7 2023

  33. [33]

    Optimal Design of Experiments for Bio-Processes using Hybrid Models

    Samuel Stricker. “Optimal Design of Experiments for Bio-Processes using Hybrid Models”. English. MA thesis. Zurich: ETH Zurich, Sept. 2024

    2024

  34. [34]

    Pareto Front Guided Sampling for Efficient Bioprocess Experimentation

    Samuel Stricker et al. “Pareto Front Guided Sampling for Efficient Bioprocess Experimentation”. en. In:System & Control Transactions. PSE Press, June 2026

    2026

  35. [35]

    A Comprehensive Review on Implementation of Quality by Design in Varied Pharmaceutical Formulations

    Aqsa Qaisar et al. “A Comprehensive Review on Implementation of Quality by Design in Varied Pharmaceutical Formulations”. en. In:Global Pharmaceutical Sciences Review9.1 (Mar. 2024), pp. 1–9.ISSN: 2788-5445.DOI: 10.31703/gpsr.2024(IX-I).01

    work page doi:10.31703/gpsr.2024(ix-i).01 2024

  36. [36]

    Carl Edward Rasmussen and Christopher K. I. Williams.Gaussian Processes for Machine Learning. en. The MIT Press, Nov. 2005.ISBN: 978-0-262-25683-4.DOI:10.7551/mitpress/3206.001.0001

    work page doi:10.7551/mitpress/3206.001.0001 2005

  37. [37]

    Hybrid Gaussian Process Models for continuous time series in bolus fed- batch cultures

    M. Nicolás Cruz-Bournazou et al. “Hybrid Gaussian Process Models for continuous time series in bolus fed- batch cultures”. In:IFAC-PapersOnLine. 13th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems DYCOPS 2022 55.7 (Jan. 2022), pp. 204–209.ISSN: 2405-8963.DOI: 10 . 1016 / j . ifacol.2022.07.445

    2022

  38. [38]

    Bayesian Optimization in Bioprocess Engineering—Where Do We Stand Today?

    Florian Gisperg et al. “Bayesian Optimization in Bioprocess Engineering—Where Do We Stand Today?” en. In: Biotechnology and Bioengineering122.6 (2025), pp. 1313–1325.ISSN: 1097-0290.DOI: 10.1002/bit.28960

    work page doi:10.1002/bit.28960 2025

  39. [39]

    doi:10.1109/TIT.2011.2182033

    Niranjan Srinivas et al. “Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting”. In:IEEE Transactions on Information Theory58.5 (May 2012), pp. 3250–3265.ISSN: 1557-9654.DOI: 10.1109/TIT.2011.2182033

    work page doi:10.1109/tit.2011.2182033 2012

  40. [40]

    Maximilian Balandat et al.BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization. Dec. 2020

    2020

  41. [41]

    Batch Bayesian Optimization

    Nathan Hunt. “Batch Bayesian Optimization”. PhD thesis. Massachusetts Institute of Technology, 2020

    2020

  42. [42]

    Bayesian Optimization with Unknown Constraints

    Michael A. Gelbart, Jasper Snoek, and Ryan P. Adams.Bayesian Optimization with Unknown Constraints. Mar. 2014.DOI:10.48550/arXiv.1403.5607

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1403.5607 2014

  43. [43]

    On Safety in Safe Bayesian Optimization

    Christian Fiedler et al. “On Safety in Safe Bayesian Optimization”. en. In:Transactions on Machine Learning Research(May 2024).ISSN: 2835-8856

    2024

  44. [44]

    Pymoo: Multi-Objective Optimization in Python

    Julian Blank and Kalyanmoy Deb. “Pymoo: Multi-Objective Optimization in Python”. In:IEEE Access8 (2020), pp. 89497–89509.ISSN: 2169-3536

    2020

  45. [45]

    Process model comparison and transferability across bioreactor scales and modes of operation for a mammalian cell bioprocess

    Stephen Craven et al. “Process model comparison and transferability across bioreactor scales and modes of operation for a mammalian cell bioprocess”. en. In:Biotechnology Progress29.1 (2013), pp. 186–196.ISSN: 1520-6033.DOI:10.1002/btpr.1664. 20 APREPRINT- JUNE18, 2026 A Appendix A.1 Dashboard Visualization Fig A1: The setting section loads the relevant r...

    work page doi:10.1002/btpr.1664 2013