Towards Autonomous Experimentation: Bayesian Optimization over Problem Formulation Space for Accelerated Alloy Development

Brent Vela; Danial Khatamsaz; Douglas L. Allaire; Joseph Wagner; Raymundo Arroyave

REVIEW 4 major objections 5 minor 47 references

Towards Autonomous Experimentation: Bayesian Optimization over Problem Formulation Space for Accelerated Alloy Development

T0 review · 4 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read The paper claims that treating the design problem itself as the search variable—via Bayesian optimization over a space of Normal Boundary Intersection formulations—lets an autonomous loop converge on an alloy that satisfies all specified…

desk verdict The core idea is from the authors' own prior work, but the alloy application and feasibility filter are solid; the paper needs a defined CHIM projection, baselines, and less inflated claims before it's publishable. read the letter →

arxiv 2502.05735 v1 pith:ZD2Q6TV7 submitted 2025-02-09 eess.SY cs.CEcs.LGcs.SYmath.OCstat.ML

Danial Khatamsaz , Joseph Wagner , Brent Vela , Raymundo Arroyave , Douglas L. Allaire This is my paper · ORCID

classification eess.SYcs.CEcs.LGcs.SYmath.OCstat.ML

keywords AutonomousDesignBayesianOptimizationProblemFormulationSpaceNormalBoundaryIntersectionAlloyDevelopmentRefractoryHighEntropyAlloysMulti-attributeUtilityGaussianProcess

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

Alloy design campaigns usually fix the optimization problem before searching, so when experiments reveal that the objectives or thresholds were wrong, the campaign restarts from scratch. This paper argues that the design problem itself can be treated as the search variable: a Bayesian optimizer runs over a 'problem formulation space' whose points are the coefficient vectors $\beta$ of Normal Boundary Intersection subproblems, with a Gaussian process mapping each formulation to the utility of its solution. In an in silico Mo–Nb–Ti–V–W case study for gas turbine blades, the loop converges to an alloy whose Cauchy pressure, yield strength, density, and solidification range jointly maximize a fixed multi-attribute utility function. The authors' claim is that discovering the problem formulation inside the loop lets an autonomous campaign adapt to evolving preferences without discarding information from previously solved problems.

What carries the argument

The central object is the problem formulation space, parametrized by vectors $\beta$ on the convex hull of individual minima (CHIM), the convex surface whose vertices are the individual minima of the quantities of interest. The load-bearing machinery is the Normal Boundary Intersection (NBI) method: each $\beta$ specifies a subproblem that is solved by maximizing the distance $c$ along a quasi-normal direction until the objective surface is reached, and the paper makes $\beta$ itself the decision variable. Distances between formulations are measured by the $\ell^1$ norm of $\beta$ differences, which feeds a squared-exponential kernel for a Gaussian process over formulations; a second Gaussian process paired with a binary classifier flags formulations that have no feasible solution, and kernel density estimation generates new candidate points. The identity doing the work is that varying $\beta$ varies the design problem, so optimizing over $\beta$ is equivalent to letting the system select which problem to solve next.

What would settle it

Enumerate the full $5$ at.$\%$ factorial grid used to bound the utility functions, compute the multi-attribute utility of every composition with the same property models, and check that the reported optimum (Cauchy pressure $91.4$ GPa, yield strength $206.6$ MPa, density $7.8$ g/cc, solidification range $44.7$ K) is the global maximizer; any grid composition with higher $U$ would falsify the convergence claim. A second check: solve the NBI subproblem at the final $\beta$ and confirm that its solution reproduces those four properties, since otherwise the projection or surrogate is mislabeling the training data.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that Bayesian optimization can be conducted over the space of design problems rather than only over the space of designs. Each point of this space is a vector $\beta$ on the convex hull of individual minima (CHIM), and together with the quasi-normal direction it defines a Normal Boundary Intersection subproblem whose solution is a candidate alloy. A Gaussian process maps $\beta$ to the multi-attribute utility of that candidate, Expected Improvement chooses the next $\beta$ to solve, and a paired classifier and regression model filter out formulations whose quasi-normal line never meets the objective space. In the demonstration, 30 replications of 40 iterations converge to a formulation whose solution has Cauchy pressure $91.4$ GPa, yield strength $206.6$ MPa, density $7.8$ g/cc, and solidification range $44.7$ K, which the paper identifies as the 'sweet spot' that maximizes $U = 1.5\,u_{\mathrm{cp}} + 1.3\,u_{\mathrm{ys}} + u_{\rho} + u_{\mathrm{sr}}$. The conclusion is that solving the problem-formulation discovery problem inside the loop is what allows the campaign to meet critical performance thresholds without an exact problem definition at the outset.

Load-bearing premise

The load-bearing premise is that projecting an alloy's property values onto the convex hull of individual minima yields a unique $\beta$ vector that faithfully represents the design problem whose solution would produce that alloy; the paper does not specify how this projection is chosen even though the equation $q = \Phi\beta + c\hat{n}$ leaves it ambiguous.

Editorial extensions

If this is right

A design campaign can change its objectives or thresholds mid-course without discarding earlier work, because solved problems remain encoded as points in formulation space and the surrogate is rebuilt rather than the campaign restarted.
The intended way to supply preferences is pairwise A/B comparison or ranking, so a decision-maker could guide exploration by choosing between designs instead of writing a scalar objective.
The two-stage feasibility filter (classifier plus regression GP) addresses the NBI defect that some subproblems have no solution, and the paper shows that false predictions decrease as the loop learns.
In the demonstration the discovered formulation yields an alloy that meets the stated thresholds for ductility, yield strength, density, and solidification range, namely the composition listed in Table 2.

Reading between the lines

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

Since the projection of an alloy's properties onto the CHIM is not unique, the method's reproducibility depends on the unstated projection rule; a natural check is whether alternative valid projections of the same 40 starting alloys lead to the same final optimum.
The framework assumes that similar $\beta$ vectors imply similar utilities via the $\ell^1$ kernel; testing that assumption with other kernels or distance metrics would show how sensitive the result is to the notion of problem similarity.
The demonstration replaces human preferences with a fixed utility function, so the practical question of noisy or inconsistent A/B feedback remains open; a simulated noisy-vote experiment would bound how many extra iterations are needed to reach the same utility.
The NBI-based formulation space is generic, so the same loop could be applied to other multi-objective design problems (electrolytes, processing parameters) without changing the algorithm, though the paper demonstrates only one alloy system.

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.

Desk Editor's Note

read the letter

Read this for the worked example, not for the new paradigm. The idea of doing Bayesian optimization over the NBI coefficient space (beta) comes from the authors' previous paper (Ref [45]); citing it only for GP background undersells that. What's actually new here is the application to a real-ish Mo-Nb-Ti-V-W alloy problem and the feasibility classifier that filters NBI subproblems with no solution — a practical issue that the NBI literature often ignores.

The demonstration is decent: 40 initial alloys, 40 iterations, 30 replications, property models that are established (Curtin-Maresca, Thermo-Calc), and a clear description of the loop. The authors are also honest about the classifier's early failures. As a proof-of-concept that you can search problem formulations rather than compositions, it works.

But there are three soft spots, and one is load-bearing. First, the projection of initial alloys onto the CHIM is underspecified. For a property vector q, q = Phi*beta + c*n_hat = Phi*(beta + c*1) is non-unique because n_hat lies in the column space of Phi. The paper says 'linear algebra' but doesn't give the rule. If the beta you assign to an alloy doesn't correspond to an actual NBI subproblem, the GP utility surrogate is trained on mismatched pairs. That matters because those 40 points are the only initialization. Fix with an explicit projection (e.g., least squares with the L1 constraint) and a sanity check that the beta maps back to the original q.

Second, no baselines. The 'efficiency' claim needs a comparison: random search in beta space, or standard BO over composition, or multi-objective BO. The average utility plot shows convergence, but not that this is faster than anything else.

Third, the autonomy claim is oversold. The utility function in Eq. 7 is hand-built with weights and thresholds; the algorithm just maximizes it. That's fine for a demo, but it's not 'without requiring an exact problem definition'—the utility is the problem definition. The A/B testing discussion is speculative, and the paper doesn't show how pairwise preferences would be converted into the GP target.

One small observation: the optimal alloy has 0% W, so the 'HEA' is really quaternary. Not a flaw, but worth noting in the discussion.

Verdict: The paper deserves a serious referee, because the application is concrete and the feasibility-filter idea is worth recording. But the referee should require the projection rule, a baseline comparison, and toned-down claims. As it stands, it's a conditional accept at best.

Who reads it: researchers in self-driving labs/materials informatics who want a worked example of problem-formulation search. I'd bring it to a reading group.

Referee Report

4 major / 5 minor

Summary. The paper proposes a framework for autonomous materials design in which Bayesian optimization is performed over a problem formulation space spanned by normal boundary intersection (NBI) subproblem coefficients (β-vectors). The central idea is that instead of specifying a fixed optimization problem, the system searches over possible formulations, mapping each to a utility value that encodes decision-maker preferences. The method is demonstrated in silico for a Mo-Nb-Ti-V-W refractory alloy system, with four quantities of interest: Cauchy pressure, yield strength, density, and solidification range. The authors construct a multi-attribute utility function, initialize a Gaussian process (GP) by projecting 40 existing alloys onto the convex hull of individual minima (CHIM), and then iteratively propose new formulations using expected improvement, with a classifier and regression model to filter infeasible formulations. The paper reports that the framework converges to a composition satisfying all thresholds and maximizing the utility function.

Significance. If the approach works as claimed, it addresses a real gap in autonomous experimentation: most closed-loop design methods assume a fixed problem formulation, whereas real campaigns often need to reformulate objectives as data arrive. The paper introduces a clear mathematical structure for the problem formulation space and demonstrates a concrete algorithmic pipeline. It also includes useful ideas such as handling infeasible NBI subproblems via a trained classifier. The authors are transparent about the limitations of their demonstration, noting that the utility function is a static stand-in for human preferences. However, the significance is tempered by the lack of any comparative baseline, an underspecified initialization step that may corrupt the GP training data, and an overstated claim of autonomy given that the utility function fully specifies the design problem in advance. The framework's potential is real, but the current evidence is not sufficient to establish that it outperforms simpler strategies such as direct Bayesian optimization over the alloy composition space or random sampling in β-space.

major comments (4)

[Section 2.3 (initialization) and Eq. (2)] End of major comment 1.
[Section 3.3 (Results) and Fig. 4] End of major comment 2.
[Abstract, Section 1.3, and Section 3.2] End of major comment 3.
[Eq. (6) and Section 2.2 (kernel definition)] End of major comment 4.

minor comments (5)

[Eq. (1) and Eq. (3)] End of minor comment 1.
[Section 2.2, definition of Φ] End of minor comment 2.
[Fig. 4 and surrounding text] End of minor comment 3.
[Fig. 7 and Fig. 8 captions] End of minor comment 4.
[Section 3.3, 'ground truth' properties] End of minor comment 5.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the framework optimizes a stated utility function and the main result is a demonstration, not a preference-discovery claim.

full rationale

The paper's derivation chain is self-contained. The utility function U (Eq. 7) is an explicitly stated input encoding fixed preferences for an in-silico demonstration, not an output derived from the algorithm; the 'sweet spot' is the maximizer of this pre-specified objective, which is the normal result of an optimization, not a circular reduction. The problem-formulation space is built on the external Normal Boundary Intersection method of Das and Dennis with the CHIM/n-hat construction, and the property models (Cauchy pressure, Curtin-Maresca strength, Thermo-Calc) are external. The Bayesian optimization loop performs a genuine search over beta, with acquisition (EI) and feasibility filtering, so the reported optimum is not forced by the initial 40-point projection by construction. Some self-citations (e.g., [45] for GP-over-formulation-space, [2-4] for alloy campaign context) provide background but are not load-bearing; the core method is defined in the paper and validated against external property models. The underspecified CHIM projection of initial alloys is a correctness/identifiability risk (the utility label may be attached to a beta that is not the solution of Eq. 2), but it is an implementation flaw and does not make the claimed result equal to its inputs by definition.

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

The framework introduces no new physical entities. The central result rests on hand-chosen utility parameters, a smoothness assumption for the GP, the accuracy of external property models, and an underspecified CHIM projection, none of which are independently validated.

free parameters (5)

Multi-attribute utility weights (1.5, 1.3, 1, 1) = 1.5, 1.3, 1, 1
Hand-chosen in Eq. 7 to reflect priorities in real alloy campaigns; the optimal alloy directly depends on these weights.
Cauchy pressure utility break point = 70 GPa
Piecewise exponential-linear utility in Fig. 5; break at 70 GPa chosen without sensitivity analysis.
Yield strength sigmoid critical point = 200 MPa (99% utility)
Sigmoid utility with critical point at 200 MPa, chosen based on expert opinion.
Density sigmoid inflection point = 9 g/cc with floor at 8 g/cc
Density utility half-max at 9 g/cc, no utility below 8 g/cc; hand-specified.
GP length scale l = not reported
Squared exponential kernel in Eq. 6; length scale is a free hyperparameter whose value or fitting procedure is not stated.

assumptions (5)

standard math NBI subproblems yield points on the Pareto front when a solution exists
Inherited from Das and Dennis (Refs. [41,42]); the paper relies on this to map beta to designs.
ad hoc to paper L1 distance between beta vectors is a meaningful dissimilarity measure between problem formulations
Eq. 4 labels this an optimal transport cost, but it is just an L1 norm; no justification that this metric reflects decision-maker relevance.
domain assumption Utility is a smooth function of beta so a squared exponential GP is appropriate
The GP in Eq. 5-6 assumes smoothness of U over beta; no verification.
domain assumption Curtin-Maresca model and Thermo-Calc predictions are accurate ground truth for the in silico alloy properties
Section 3.1 uses these models as ground truth; errors would propagate to utilities.
ad hoc to paper The projection of initial data onto CHIM is unique and defines valid problem formulations
Section 2.3 initialization; not specified and generally underdetermined.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Towards Autonomous Experimentation: Bayesian Optimization over Problem Formulation Space for Accelerated Alloy Development." pith.science (2026). https://pith.science/paper/ZD2Q6TV7

@misc{pith2026250205735,
  author       = {Pith},
  title        = {Pith review of: Towards Autonomous Experimentation: Bayesian Optimization over Problem Formulation Space for Accelerated Alloy Development},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZD2Q6TV7}},
  note         = {Machine review of arXiv:2502.05735}
}

read the original abstract

Accelerated discovery in materials science demands autonomous systems capable of dynamically formulating and solving design problems. In this work, we introduce a novel framework that leverages Bayesian optimization over a problem formulation space to identify optimal design formulations in line with decision-maker preferences. By mapping various design scenarios to a multi attribute utility function, our approach enables the system to balance conflicting objectives such as ductility, yield strength, density, and solidification range without requiring an exact problem definition at the outset. We demonstrate the efficacy of our method through an in silico case study on a Mo-Nb-Ti-V-W alloy system targeted for gas turbine engine blade applications. The framework converges on a sweet spot that satisfies critical performance thresholds, illustrating that integrating problem formulation discovery into the autonomous design loop can significantly streamline the experimental process. Future work will incorporate human feedback to further enhance the adaptability of the system in real-world experimental settings.

Figures

Figures reproduced from arXiv: 2502.05735 by the authors.

**Figure 1.** Schematic of a real development campaign to identify refractory alloys for ultra-high temperature applications up to 2000 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

**Figure 2.** Illustration of the normal boundary intersection method for discov [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Step-by-step representation of our proposed framework. After completing the initialization steps, the Bayesian design loop explores the problem formu [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Comparison of optimum utility versus the average achieved utility [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

**Figure 5.** Figure 5: Utility value functions constructed to capture the decision-maker’s [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

**Figure 7.** Figure 7: Average number of failed attempts from 30 simulations. Early iter [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]

**Figure 6.** Figure 6: Affine projection of the Mo-Nb-Ti-V-W composition space into 2D. The vertices represent pure elements, and the color scale indicates the normalized utility values. In contrast to human-in-the-loop scenarios—where Bayesian optimization directly explores the alloy space…

**Figure 8.** Figure 8: Kernel density estimates of the distribution of feasible problem for [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 41 canonical work pages

[45]

Wagner, D

J. Wagner, D. Khatamsaz, D. L. Allaire, Semi-autonomous problem for- mulation space search for high dimensional multiobjective optimization, AIAA A VIATION 2023 Forum (2023). URL https://api.semanticscholar.org/CorpusID:259824275

work page 2023
[1]

Arr ´oyave, D

R. Arr ´oyave, D. Khatamsaz, B. Vela, R. Couperthwaite, A. Molkeri, D. D. Singh Prashant and Johnson, X. Qian, D. Srivastava Ankit and Allaire, A perspective on Bayesian methods applied to materials dis- covery and design, MRS Commun. 12 (6) (2022) 1037–1049

work page 2022
[2]

Acemi, B

C. Acemi, B. Vela, E. Norris, W. Trehern, K. C. Atli, C. Cleek, R. Ar- royave, I. Karaman, Multi-objective, multi-constraint high-throughput design, synthesis, and characterization of tungsten-containing refractory multi-principal element alloys, Acta Materialia (2024) 120379

work page 2024
[3]

Accelerated Multi-Objective Alloy Discovery through Efficient Bayesian Methods: Application to the FCC Alloy Space

T. Hastings, M. Mulukutla, D. Khatamsaz, D. Salas, W. Xu, D. Lewis, N. Person, M. Skokan, B. Miller, J. Paramore, et al., An interoperable multi objective batch bayesian optimization framework for high through- put materials discovery, arXiv preprint arXiv:2405.08900 (2024)

work page Pith review arXiv 2024
[4]

Mulukutla, A

M. Mulukutla, A. N. Person, S. V oigt, L. Kuettner, B. Kappes, D. Khatamsaz, R. Robinson, D. S. Mula, W. Xu, D. Lewis, et al., Illustrat- ing an effective workflow for accelerated materials discovery, Integrating Materials and Manufacturing Innovation (2024) 1–21

work page 2024
[5]

Y . Shi, P. L. Prieto, T. Zepel, S. Grunert, J. E. Hein, Automated experi- mentation powers data science in chemistry, Accounts of Chemical Re- search 54 (3) (2021) 546–555

work page 2021
[6]

Crabtree, Self-driving laboratories coming of age, Joule 4 (12) (2020) 2538–2541

G. Crabtree, Self-driving laboratories coming of age, Joule 4 (12) (2020) 2538–2541. doi:https://doi.org/10.1016/j.joule.2020.11. 021. URL https://www.sciencedirect.com/science/article/pii/ S2542435120305675

work page doi:10.1016/j.joule.2020.11 2020
[7]

Ament, M

S. Ament, M. Amsler, D. R. Sutherland, M.-C. Chang, D. Guevarra, A. B. Connolly, J. M. Gregoire, M. O. Thompson, C. P. Gomes, R. B. van Dover, Autonomous materials synthesis via hierarchical active learn- ing of nonequilibrium phase diagrams, Science Advances 7 (51) (2021) eabg4930

work page 2021

Show all 47 references

[8]

D. Bash, Y . Cai, V . Chellappan, S. L. Wong, X. Yang, P. Kumar, J. D. Tan, A. Abutaha, J. J. Cheng, Y .-F. Lim, et al., Multi-fidelity high-throughput optimization of electrical conductivity in p3ht-cnt composites, Advanced Functional Materials 31 (36) (2021) 2102606

2021
[9]

Langner, F

S. Langner, F. H ¨ase, J. D. Perea, T. Stubhan, J. Hauch, L. M. Roch, T. Heumueller, A. Aspuru-Guzik, C. J. Brabec, Beyond ternary opv: high- throughput experimentation and self-driving laboratories optimize multi- component systems, Advanced Materials 32 (14) (2020) 1907801

2020
[10]

A. E. Gongora, B. Xu, W. Perry, C. Okoye, P. Riley, K. G. Reyes, E. F. Morgan, K. A. Brown, A bayesian experimental autonomous researcher for mechanical design, Science advances 6 (15) (2020) eaaz1708

2020
[11]

J. R. Deneault, J. Chang, J. Myung, D. Hooper, A. Armstrong, M. Pitt, B. Maruyama, Toward autonomous additive manufacturing: Bayesian op- timization on a 3d printer, MRS Bulletin 46 (7) (2021) 566–575

2021
[12]

R. W. Epps, M. S. Bowen, A. A. V olk, K. Abdel-Latif, S. Han, K. G. Reyes, A. Amassian, M. Abolhasani, Artificial chemist: an autonomous quantum dot synthesis bot, Advanced Materials 32 (30) (2020) 2001626

2020
[13]

Christensen, L

M. Christensen, L. P. Yunker, F. Adedeji, F. H¨ase, L. M. Roch, T. Gensch, G. dos Passos Gomes, T. Zepel, M. S. Sigman, A. Aspuru-Guzik, et al., Data-science driven autonomous process optimization, Communications Chemistry 4 (1) (2021) 1–12

2021
[14]

Grizou, L

J. Grizou, L. J. Points, A. Sharma, L. Cronin, A curious formulation robot enables the discovery of a novel protocell behavior, Science advances 6 (5) (2020) eaay4237

2020
[15]

L. Cao, D. Russo, K. Felton, D. Salley, A. Sharma, G. Keenan, W. Mauer, H. Gao, L. Cronin, A. A. Lapkin, Optimization of formulations using robotic experiments driven by machine learning doe, Cell Reports Physi- cal Science 2 (1) (2021) 100295. 10

2021
[16]

N. J. Szymanski, Y . Zeng, H. Huo, C. J. Bartel, H. Kim, G. Ceder, Toward autonomous design and synthesis of novel inorganic materials, Materials Horizons 8 (8) (2021) 2169–2198

2021
[17]

B. P. MacLeod, F. G. Parlane, C. C. Rupnow, K. E. Dettelbach, M. S. El- liott, T. D. Morrissey, T. H. Haley, O. Proskurin, M. B. Rooney, N. Taher- imakhsousi, et al., A self-driving laboratory advances the pareto front for material properties, Nature communications 13 (1) (2022) 1–10

2022
[18]

H ¨ase, L

F. H ¨ase, L. M. Roch, A. Aspuru-Guzik, Next-generation experimentation with self-driving laboratories, Trends in Chemistry 1 (3) (2019) 282–291

2019
[19]

B. P. MacLeod, F. G. Parlane, K. E. Dettelbach, M. S. Elliott, C. C. Rup- now, T. D. Morrissey, T. H. Haley, O. Proskurin, M. B. Rooney, N. Taher- imakhsousi, et al., Advancing the pareto front using a self-driving labora- tory, arXiv preprint arXiv:2106.08899 (2021)

2021 arXiv
[20]

K. Deb, Multi-objective optimisation using evolutionary algorithms: an introduction, in: Multi-objective evolutionary optimisation for product design and manufacturing, Springer, 2011, pp. 3–34

2011
[21]

Konak, D

A. Konak, D. W. Coit, A. E. Smith, Multi-objective optimization using genetic algorithms: A tutorial, Reliability engineering & system safety 91 (9) (2006) 992–1007

2006
[22]

Mingqiang, K

L. Mingqiang, K. Jisong, D. Lin, Ga-based multi-objective optimization, in: Proceedings of the 3rd World Congress on Intelligent Control and Automation (Cat. No. 00EX393), V ol. 1, IEEE, 2000, pp. 637–640

2000
[23]

Nedjah, L

N. Nedjah, L. d. M. Mourelle, Evolutionary multi–objective optimisation: A survey, International Journal of Bio-Inspired Computation 7 (1) (2015) 1–25

2015
[24]

Khatamsaz, L

D. Khatamsaz, L. Peddareddygari, S. Friedman, D. Allaire, Bayesian op- timization of multiobjective functions using multiple information sources, AIAA Journal 59 (6) (2021) 1964–1974

2021
[25]

Arr ´oyave, D

R. Arr ´oyave, D. Khatamsaz, B. Vela, R. Couperthwaite, A. Molkeri, P. Singh, D. D. Johnson, X. Qian, A. Srivastava, D. Allaire, A perspec- tive on bayesian methods applied to materials discovery and design, MRS communications 12 (6) (2022) 1037–1049

2022
[26]

S. M. A. A. Alvi, J. Janssen, D. Khatamsaz, D. Perez, D. Allaire, R. Arroyave, Hierarchical gaussian process-based bayesian optimiza- tion for materials discovery in high entropy alloy spaces, arXiv preprint arXiv:2410.04314 (2024)

2024 arXiv
[27]

J. Knowles, Parego: A hybrid algorithm with on-line landscape approxi- mation for expensive multiobjective optimization problems, IEEE trans- actions on evolutionary computation 10 (1) (2006) 50–66

2006
[28]

Hakanen, J

J. Hakanen, J. D. Knowles, On using decision maker preferences with parego, in: Evolutionary Multi-Criterion Optimization: 9th International Conference, EMO 2017, M ¨unster, Germany, March 19-22, 2017, Pro- ceedings 9, Springer, 2017, pp. 282–297

2017
[29]

URL https://arpa-e.energy.gov/technologies/programs/ ultimate

Ultrahigh temperature impervious materials advancing turbine e fficiency (ultimate) (Nov 2020). URL https://arpa-e.energy.gov/technologies/programs/ ultimate

2020
[30]

Khatamsaz, B

D. Khatamsaz, B. Vela, P. Singh, D. D. Johnson, D. Allaire, R. Arr ´oyave, Multi-objective materials bayesian optimization with active learning of design constraints: Design of ductile refractory multi-principal-element alloys, Acta Materialia 236 (2022) 118133

2022
[31]

Khatamsaz, B

D. Khatamsaz, B. Vela, P. Singh, D. D. Johnson, D. Allaire, R. Arr ´oyave, Bayesian optimization with active learning of design constraints using an entropy-based approach, npj Computational Materials 9 (1) (2023) 49

2023
[32]

Beume, S-Metric Calculation by Considering Dominated Hypervol- ume as Klee’s Measure Problem, Evolutionary Computation 17 (4) (2009) 477–492

N. Beume, S-Metric Calculation by Considering Dominated Hypervol- ume as Klee’s Measure Problem, Evolutionary Computation 17 (4) (2009) 477–492. doi:10.1162/evco.2009.17.4.17402. URL https://doi.org/10.1162/evco.2009.17.4.17402

2009 doi
[33]

Bradstreet, L

L. Bradstreet, L. While, L. Barone, A fast many-objective hypervolume algorithm using iterated incremental calculations, in: IEEE Congress on Evolutionary Computation, 2010, pp. 1–8. doi:10.1109/CEC.2010. 5586344

2010 doi
[34]

M. T. M. Emmerich, A. H. Deutz, J. W. Klinkenberg, Hypervolume-based expected improvement: Monotonicity properties and exact computation, in: 2011 IEEE Congress of Evolutionary Computation (CEC), 2011, pp. 2147–2154. doi:10.1109/CEC.2011.5949880

2011
[35]

C. M. Fonseca, L. Paquete, M. Lopez-Ibanez, An Improved Dimension- Sweep Algorithm for the Hypervolume Indicator, in: 2006 IEEE Interna- tional Conference on Evolutionary Computation, 2006, pp. 1157–1163. doi:10.1109/CEC.2006.1688440

2006 arXiv
[36]

L. M. S. Russo, A. P. Francisco, Quick Hypervolume, IEEE Transactions on Evolutionary Computation 18 (2012) 481–502. URL https://api.semanticscholar.org/CorpusID:7504127

2012
[37]

Q. Yang, S. Ding, Novel Algorithm to Calculate Hypervolume Indicator of Pareto Approximation Set, in: International Conference on Intelligent Computing, 2007. URL https://api.semanticscholar.org/CorpusID:879

2007
[38]

Zitzler, L

E. Zitzler, L. Thiele, Multiobjective evolutionary algorithms: a compar- ative case study and the strength Pareto approach, IEEE Transactions on Evolutionary Computation 3 (4) (1999) 257–271. doi:10.1109/4235. 797969

1999 doi
[39]

Marler, J

R. Marler, J. Arora, The weighted sum method for multi-objective opti- mization: New insights, Structural and Multidisciplinary Optimization 41 (2010) 853–862. doi:10.1007/s00158-009-0460-7

2010 doi
[40]

I. Y . Kim, O. L. de Weck, Adaptive weighted-sum method for bi- objective optimization: Pareto front generation, Structural and Mul- tidisciplinary Optimization 29 (2) (2005) 149–158. doi:10.1007/ s00158-004-0465-1 . URL https://doi.org/10.1007/s00158-004-0465-1

2005 doi
[41]

Das, Nonlinear multicriteria optimization and robust optimality, Rice University, 1997

I. Das, Nonlinear multicriteria optimization and robust optimality, Rice University, 1997

1997
[42]

I. Das, J. E. Dennis, Normal-boundary intersection: A new method for generating the pareto surface in nonlinear multicriteria optimiza- tion problems, SIAM Journal on Optimization 8 (3) (1998) 631–657. arXiv:https://doi.org/10.1137/S1052623496307510, doi:10. 1137/S105262349630...

1998 doi
[43]

A. M. Vershik, Long history of the monge-kantorovich transportation problem, The Mathematical Intelligencer 35 (4) (2013) 1–9

2013
[44]

Villani, Optimal transport: old and new, V ol

C. Villani, Optimal transport: old and new, V ol. 338, Springer, 2009

2009
[46]

Maresca, W

F. Maresca, W. A. Curtin, Mechanistic origin of high strength in refrac- tory bcc high entropy alloys up to 1900k, Acta Mater. 182 (2020) 235–

2020
[249]

URL https://www.sciencedirect.com/science/article/pii/ S1359645419306755 11

doi:https://doi.org/10.1016/j.actamat.2019.10.015. URL https://www.sciencedirect.com/science/article/pii/ S1359645419306755 11

2019 doi

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [45]

Wagner, D

J. Wagner, D. Khatamsaz, D. L. Allaire, Semi-autonomous problem for- mulation space search for high dimensional multiobjective optimization, AIAA A VIATION 2023 Forum (2023). URL https://api.semanticscholar.org/CorpusID:259824275

work page 2023

[2] [1]

Arr ´oyave, D

R. Arr ´oyave, D. Khatamsaz, B. Vela, R. Couperthwaite, A. Molkeri, D. D. Singh Prashant and Johnson, X. Qian, D. Srivastava Ankit and Allaire, A perspective on Bayesian methods applied to materials dis- covery and design, MRS Commun. 12 (6) (2022) 1037–1049

work page 2022

[3] [2]

Acemi, B

C. Acemi, B. Vela, E. Norris, W. Trehern, K. C. Atli, C. Cleek, R. Ar- royave, I. Karaman, Multi-objective, multi-constraint high-throughput design, synthesis, and characterization of tungsten-containing refractory multi-principal element alloys, Acta Materialia (2024) 120379

work page 2024

[4] [3]

Accelerated Multi-Objective Alloy Discovery through Efficient Bayesian Methods: Application to the FCC Alloy Space

T. Hastings, M. Mulukutla, D. Khatamsaz, D. Salas, W. Xu, D. Lewis, N. Person, M. Skokan, B. Miller, J. Paramore, et al., An interoperable multi objective batch bayesian optimization framework for high through- put materials discovery, arXiv preprint arXiv:2405.08900 (2024)

work page Pith review arXiv 2024

[5] [4]

Mulukutla, A

M. Mulukutla, A. N. Person, S. V oigt, L. Kuettner, B. Kappes, D. Khatamsaz, R. Robinson, D. S. Mula, W. Xu, D. Lewis, et al., Illustrat- ing an effective workflow for accelerated materials discovery, Integrating Materials and Manufacturing Innovation (2024) 1–21

work page 2024

[6] [5]

Y . Shi, P. L. Prieto, T. Zepel, S. Grunert, J. E. Hein, Automated experi- mentation powers data science in chemistry, Accounts of Chemical Re- search 54 (3) (2021) 546–555

work page 2021

[7] [6]

Crabtree, Self-driving laboratories coming of age, Joule 4 (12) (2020) 2538–2541

G. Crabtree, Self-driving laboratories coming of age, Joule 4 (12) (2020) 2538–2541. doi:https://doi.org/10.1016/j.joule.2020.11. 021. URL https://www.sciencedirect.com/science/article/pii/ S2542435120305675

work page doi:10.1016/j.joule.2020.11 2020

[8] [7]

Ament, M

S. Ament, M. Amsler, D. R. Sutherland, M.-C. Chang, D. Guevarra, A. B. Connolly, J. M. Gregoire, M. O. Thompson, C. P. Gomes, R. B. van Dover, Autonomous materials synthesis via hierarchical active learn- ing of nonequilibrium phase diagrams, Science Advances 7 (51) (2021) eabg4930

work page 2021