Pith. sign in

REVIEW 4 major objections 6 minor 60 references

Generative Multi-Form Bayesian Optimization

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 ↗

Pith's one-line read GMFoO resolves the latent-dimension accuracy-versus-speed trade-off in generative optimization by running Bayesian optimization on correlated high- and low-dimensional latent spaces simultaneously.

desk verdict A genuinely new multi-latent-space GMO scheme with InfoGAN-style correlation and co-kriging transfer; the core idea is plausible and the experiments mostly support it, but the 'always best' claim overreaches the evidence. read the letter →

arxiv 2501.13337 v1 pith:TQ3M35NO submitted 2025-01-23 cs.CE

Zhendong Guo , Haitao Liu , Yew-Soon Ong , Xinghua Qu , Yuzhe Zhang , Jianmin Zheng This is my paper · ORCID
classification cs.CE
keywords generativemodelbasedoptimizationmulti-formBayesianlatentspacemulti-fidelityGaussianprocessknowledgetransferairfoildesignInfoGAN
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

Generative model based optimization (GMO) maps a complex structured design space into a continuous latent space, but choosing the latent dimension forces a trade-off: low dimensions converge fast yet cap solution quality, high dimensions are accurate but slow. This paper claims that optimizing over two correlated latent spaces at once breaks that trade-off. The proposed method, GMFoO, trains a GAN whose low-dimensional latent code is regularized to correlate with its high-dimensional latent vector, then runs Bayesian optimization in both spaces, using the low-space optimum to narrow the high-space search and treating cross-projected samples as low-fidelity data in a multi-fidelity Gaussian process. On airfoil, decorative corbel, and MNIST-area test problems, the paper reports that GMFoO reaches better designs faster than single-space GMO, randomly-subspaced NashEGO, SVD-BO, and evolutionary baselines within the same sample budget. If the claim holds, GMFoO offers a practical hedge against latent-dimension misspecification in expensive black-box design.

What carries the argument

The load-bearing object is MFoO-GAN, a generative adversarial network whose training loss adds the InfoGAN mutual-information term I(c; G(z)) so that the low-dimensional code c captures the major variability of the generated object and is correlated with the high-dimensional vector z = [c, z*]. On top of this, two exchange mechanisms carry the argument: enhanced local exploitation, which narrows the search box in Z around the current best low-space solution c_min with radius Δ, and multi-fidelity GP surrogates, where samples transformed from C into Z (by fixing z*) and from Z into C (by inverse inference through the discriminator) are treated as low-fidelity data and combined with exact expensive samples through co-kriging correlation coefficients ρ_ij. The mutual-information regularization is what makes the transferred samples trustworthy, and the multi-fidelity GPs are what convert that trust into faster convergence.

What would settle it

Train MFoO-GAN on a design dataset, then compare GMFoO against GMO-High on a problem where the learned low-dimensional space is artificially decorrelated from Z, for instance by shuffling the low-dimensional codes relative to their high-dimensional partners. If GMFoO still wins, its success does not depend on the correlation assumption; if it loses, the assumption is confirmed as load-bearing. A cheaper check is to compute the Pearson correlation between y(z) and y(c') on held-out samples and test whether GMFoO's advantage disappears below some r threshold.

Watch

Extended reading notes

Core claim

The central claim is that the apparent conflict between solution accuracy and convergence rate in GMO can be resolved by multi-form optimization over multiple latent spaces of one generative model, rather than by choosing a single latent dimension. In GMFoO, a GAN with InfoGAN-style mutual information regularization, called MFoO-GAN, produces a high-dimensional space Z and one or more low-dimensional spaces C that are positively correlated; the best point found in C is used to narrow the search region in Z around [c_min, 0], and samples projected between Z and C are fed into two multi-fidelity GPs as low-fidelity data. The paper's experiments on low-speed and subsonic airfoil design, corbel design, and binary-image area maximization show GMFoO achieving the best final solutions with faster convergence than GMO-High, GMO-Low, GMO-NashEGO, SVD-BO, GMO-CMAES, and GMO-IKEA.

Load-bearing premise

The method presupposes that the low-dimensional latent space C is correlated enough with the high-dimensional space Z that c_min lies near Z's optimum and that projected C-samples improve the multi-fidelity surrogate rather than corrupt it; the paper demonstrates one Pearson r=0.8 case and does not characterise when transfer turns negative.

Editorial extensions

If this is right

  • GMFoO's best solution and convergence rate dominate single-space GMO baselines on all three test problems within the stated budgets.
  • The multi-fidelity component alone, corresponding to Δ=0, outperforms GMO-High, and adding the c_min-guided narrowing improves further, so each transfer strategy contributes independently.
  • Correlated latent spaces transfer more effectively than randomly generated alternate subspaces, explaining why GMO-NashEGO lags behind GMFoO.
  • The gains persist when the high-dimensional latent space grows from 13 to 23 dimensions, where BO-based GMFoO overtakes evolutionary GMO-IKEA within budget.
  • The framework applies to structured design spaces with hundreds of interacting variables, including 384-variable airfoil contours, corbel curves, and 784-bit binary images, rather than only low-dimensional Euclidean test functions.

Reading between the lines

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

  • The correlation requirement suggests a cheap diagnostic: measure the Pearson correlation between y(z) and y(c') before trusting transfer; the paper reports one case with r=0.8, so users could set a threshold below which low-fidelity injection is switched off.
  • The same multi-space hedge could be built on other generative models, such as VAEs or diffusion models, by imposing analogous mutual-information or reconstruction constraints between latent subspaces; nothing in the argument is specific to GANs.
  • A natural extension is to adapt Δ and the low-dimensional dimension dL online from estimated correlation and surrogate disagreement, since the paper fixes them by default and shows sensitivity to both.
  • Because negative transfer is possible when the low-space optimum drifts away from the high-space optimum, adaptive gating or weighting of transferred samples would make the method safer on problems where the generative model's latent spaces are not well correlated.
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 proposes GMFoO (generative multi-form optimization), a framework that trains a GAN-like generative model (MFoO-GAN) to produce a high-dimensional latent space Z and a low-dimensional latent space C that are positively correlated via an InfoGAN-style mutual information regularizer. Bayesian optimization is then run simultaneously in both latent spaces, with two knowledge-transfer mechanisms: a multi-fidelity Gaussian process that treats samples transformed between Z and C as low-fidelity data, and an 'enhanced local exploitation' step that narrows the Z search region around the current best point of C. Experiments on low-speed and subsonic airfoil design, decorative corbel design, and an MNIST area maximization problem compare GMFoO against GMO-High, GMO-Low, GMO-NashEGO, SVD-BO, GMO-CMAES, and GMO-IKEA, with ablations for the size parameter Δ and the low-dimensional latent dimension dL.

Significance. The paper addresses a genuine and practically relevant issue: the choice of latent dimension in generative-model-based optimization involves a trade-off between solution accuracy and convergence rate. If the empirical claims hold, GMFoO offers a useful way to hedge this choice by optimizing over multiple correlated latent spaces simultaneously, and the multi-fidelity treatment of transformed samples is an interesting idea. The manuscript includes ablations for the main algorithmic components and a correlation analysis between the latent spaces. However, the strength of the empirical claims is not yet matched by the evidence: some statements overreach the data, the comparison protocol is under-specified, and the central correlation premise is validated only for one of the four test problems.

major comments (4)
  1. [4.3, Fig. 8(b)] The claim in Section 4.3 that 'GMFoO ... always achieves the best solutions with even faster convergence rates' is contradicted later in the same section, where the authors state that for the low-speed airfoil GMO-IKEA achieves final solutions slightly better than those of GMFoO (Fig. 8(b)). Please moderate this claim and analyze the conditions under which GMFoO is and is not the best method.
  2. [4.2, Algorithm 3, Figs. 8-9] GMFoO evaluates three points per main loop iteration (one in C, one in the full Z, and one in the narrowed Z, per Algorithm 3 lines 6 and 9), while the baselines evaluate one point per iteration. If the convergence plots use iterations on the x-axis, GMFoO receives a 3x larger budget per iteration; if they use function evaluations, the number of evaluations per iteration should be stated. Please report results as a function of the total number of expensive function evaluations and describe the DoE sizes exactly.
  3. [4.6.1, Fig. 14] The two knowledge-transfer mechanisms in Section 3.2 presuppose that the low-dimensional latent space C is well correlated with Z, but the only direct evidence is the single Pearson coefficient r=0.8 for the low-speed airfoil. No correlation analysis is reported for the subsonic airfoil, corbel, or MNIST cases, and the fitted multi-fidelity GP correlation coefficients ρ_ij are not reported. Please add per-problem correlation and negative-transfer diagnostics, or explain why the single case is sufficient.
  4. [4.2, 4.6.2-4.6.3] The parameters Δ and dL are selected per problem after sensitivity analyses, and the main comparisons use the best configuration (e.g., Δ=0.15, dL=4). No statistical significance tests are applied to the 10-run comparisons, so the reported differences among algorithms may not be significant. Please report significance tests (e.g., Wilcoxon signed-rank or rank-sum tests) and state whether the sensitivity-analysis results are used to pick parameters before or after the main comparison.
minor comments (6)
  1. [Eqs. (16) and (18)] The covariance matrices written with ρ12 and ρ21 are only symmetric if ρ12 = ρ21; this constraint should be stated explicitly. Also, the shorthand K(C') and K(Z') should be defined as K(C',C') and K(Z',Z').
  2. [Eqs. (14)-(15)] The derivation treats H(c) as constant when moving from Eq. (14) to Eq. (15); this is only valid if the marginal distribution of c is fixed during generator training, which should be clarified.
  3. [Algorithm 3, line 9] It is unclear whether the point selected in the narrowed high-dimensional space is included in the training set for the multi-fidelity GP of Z; please specify the update step.
  4. [Figs. 8 and 9] The convergence-history plots are missing explicit x-axis labels; please state whether the horizontal axis is the number of iterations or the number of expensive function evaluations.
  5. [4.2] The sentence 'The number of initial training samples for the standard BO and IKEA are set as 11 times of the dimension of the low-dimensional latent space of MFoO-GAN (i.e., 11 dL)' is surprising because GMO-High optimizes in a 13- or 23-dimensional space but receives only 33 or 44 initial samples; please clarify why the DoE size is tied to dL rather than the dimension being optimized.
  6. [1] In the contribution list, 'we instantiates the proposed GMFoO' should be 'we instantiate the proposed GMFoO'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: GMFoO's central claim is an empirically benchmarked algorithm comparison, and the paper's design components do not reduce by construction to their own outputs.

full rationale

The central claim of GMFoO is an empirical algorithm-performance claim: simultaneous optimization over a high-dimensional latent space Z and a correlated low-dimensional latent space C, with continuous information exchange, yields better solutions with faster convergence on airfoil, corbel, and MNIST area-maximization benchmarks. This claim is evaluated by held-out comparison against GMO-High, GMO-Low, GMO-NashEGO, SVD-BO, GMO-CMAES, and GMO-IKEA, using convergence histories and final-solution distributions over multiple runs. The two exchange mechanisms, enhanced local exploitation and multi-fidelity GP, are implemented as described in Eqs. (16)-(21), and their individual contributions are tested by ablations (GMFoO 0 versus GMFoO 10/15/20, and variations of dL). The correlation between C and Z is promoted by construction through the InfoGAN-style regularizer and the structural choice that C is a subset of Z; this is a design mechanism, not a derived result. The paper's Pearson r=0.8 evidence for the low-speed airfoil is an empirical observation about the learned latent spaces, not an equation that forces GMFoO's advantage. The only self-citations are background references for multi-fidelity surrogates, Bayesian optimization, and the corbel dataset; none is used as a substitute for the claimed result. The sensitivity of parameters such as Δ and dL is investigated rather than hidden. Therefore, no circular step can be exhibited from the paper's equations or citations, and the derivation chain is self-contained with respect to the stated experimental protocol.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the GMO embedding assumption (from prior work), plus a new assumption that correlated low-dimensional latent spaces can be used for local exploitation and as low-fidelity sources. No code or data is shipped, and several free parameters (Delta, dL, lambda_i, rho_ij) are not fully specified or are tuned per problem.

free parameters (5)
  • Delta (narrowed search range) = 0.15 (default; sensitivity analysis over 0, 0.1, 0.15, 0.2)
    Controls the size of the local exploitation region in the high-dimensional latent space around c_min. It is set by hand and analyzed post hoc; it directly affects the high-dimensional search.
  • dL (low-dimensional latent dimension) = 3, 4, or 5 depending on problem (low-speed airfoil: 3; subsonic airfoil: 4; corbel: 5; MNIST: 4)
    Dimension of the low-dimensional latent space C. Chosen per test case; the paper concludes dL=4 performs best overall.
  • lambda_i (InfoGAN regularization weights) = not specified
    Weights of the mutual information terms in Eq.13; values are not reported, so the training loss is not fully specified.
  • rho_ij (multi-fidelity GP correlation coefficients) = not reported values; estimated via maximum likelihood
    Correlations between the high- and low-fidelity models in Eqs.16-19. These are fitted to the data during surrogate building, standard for co-kriging, but unstated values make replication harder.
  • sigma_n (GP noise terms) = not specified
    Noise variances in the multi-fidelity GP; fitted from data.
assumptions (4)
  • domain assumption The generator g maps the structured input space X to a continuous latent space Z, and optimizing over Z is a faithful proxy for optimizing over X.
    This is the standard GMO assumption, adopted from cited GMO literature (Section 2.1, Eq.1). The paper does not prove that the latent manifold contains the true optimum.
  • domain assumption The low-dimensional latent space C is positively correlated with Z and captures major variability, so that its optimum c_min is near the optimal solution in Z.
    Core to the local exploitation strategy (Section 3.2, Fig.4). The paper verifies this on one airfoil case with Pearson r=0.8 (Section 4.6.1), but it is not guaranteed for other problems.
  • ad hoc to paper The inverse inference c' = E[Q(c|x)] from the discriminator produces a point in C that can carry the label y(z) as a low-fidelity observation.
    The paper introduces this mapping (Section 3.1, Fig.3b) and relies on it for the multi-fidelity GP. It is not a standard co-kriging setup: the low-fidelity values are exact expensive evaluations placed at a projected input.
  • domain assumption Co-kriging with these transformed samples improves the GP surrogate more than it hurts it (positive transfer).
    The efficiency of the knowledge transfer in Section 3.2 depends on the co-kriging model correctly fusing exact labels at approximate inputs; no analysis of negative transfer is given.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generative Multi-Form Bayesian Optimization." pith.science (2026). https://pith.science/paper/TQ3M35NO

@misc{pith2026250113337,
  author       = {Pith},
  title        = {Pith review of: Generative Multi-Form Bayesian Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TQ3M35NO}},
  note         = {Machine review of arXiv:2501.13337}
}
read the original abstract

Many real-world problems, such as airfoil design, involve optimizing a black-box expensive objective function over complex structured input space (e.g., discrete space or non-Euclidean space). By mapping the complex structured input space into a latent space of dozens of variables, a two-stage procedure labeled as generative model based optimization (GMO) in this paper, shows promise in solving such problems. However, the latent dimension of GMO is hard to determine, which may trigger the conflicting issue between desirable solution accuracy and convergence rate. To address the above issue, we propose a multi-form GMO approach, namely generative multi-form optimization (GMFoO), which conducts optimization over multiple latent spaces simultaneously to complement each other. More specifically, we devise a generative model which promotes positive correlation between latent spaces to facilitate effective knowledge transfer in GMFoO. And further, by using Bayesian optimization (BO) as the optimizer, we propose two strategies to exchange information between these latent spaces continuously. Experimental results are presented on airfoil and corbel design problems and an area maximization problem as well to demonstrate that our proposed GMFoO converges to better designs on a limited computational budget.

Figures

Figures reproduced from arXiv: 2501.13337 by the authors.

Figure 1
Figure 1. Standard process of generative model based optimization (GMO) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. GMFoO framework, where the generative model is labeled as MFoO-GAN [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Sample synthesis and exchange process between the high- and lower-dimensional latent space, [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (15 more)
Figure 4
Figure 4. Figure 4: Enhanced local exploitation of the high-dimensional latent space by leveraging information of [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Leveraging knowledge through the multi-fidelity modeling, (a) in the high-dimensional latent [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Flowchart of GMFoO (2) To illustrate the advantage of GMFoO over traditional MFoO method, we compare it with a GMO which uses a multi-form BO algorithm (labeled as NashEGO [47]) to optimize the high-dimensional latent space of MFoO-GAN. The related procedure is labeled…
Figure 7
Figure 7. Figure 7: Introduction of (a) the airfoil contour and the forces acting on it, and (b) airfoils in UIUC [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: Testing results of low-speed airfoil, where the dimension of high- and low-dimensional latent [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Testing results of subsonic airfoil, where the dimension of high- and low-dimensional latent [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Introduction to the decorative corbel, (a) decorative corbel and its components, (b) forces [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]
Figure 11
Figure 11. Figure 11: Testing results of corbel optimization, where the dimension of high- and low-dimensional [PITH_FULL_IMAGE:figures/full_fig_p014_11.png]
Figure 12
Figure 12. Figure 12: Generated hand-writing digits by (a) the 20-dimensional latent space and (b) the 4- [PITH_FULL_IMAGE:figures/full_fig_p015_12.png]
Figure 13
Figure 13. Figure 13: Testing results of area maximization problem based on the MNIST dataset, where the dimen [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]
Figure 14
Figure 14. Figure 14: Functional correlation between high- and low-dimensional latent spaces of MFoO-GAN, by [PITH_FULL_IMAGE:figures/full_fig_p017_14.png]
Figure 15
Figure 15. Figure 15: Distributions of the optimal solutions in the high- and low-dimensional latent spaces of [PITH_FULL_IMAGE:figures/full_fig_p017_15.png]
Figure 16
Figure 16. Figure 16: Experimental results with different ∆ in GMFoO, (a) low-speed airfoil optimization, (b) [PITH_FULL_IMAGE:figures/full_fig_p018_16.png]
Figure 17
Figure 17. Figure 17: Effects of dL on GMFoO performance by testing on low-speed airfoil design, (a) dL = 3, (b) dL = 4, (c) dL = 5 [PITH_FULL_IMAGE:figures/full_fig_p019_17.png]
Figure 18
Figure 18. Figure 18: Effects of dL on GMFoO performance by testing on subsonic airfoil design, (a) dL = 3, (b) dL = 4, (c) dL = 5 19 [PITH_FULL_IMAGE:figures/full_fig_p019_18.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 47 canonical work pages

  1. [1]

    Efficient aerodynamic shape optimization with deep- learning-based geometric filtering,

    J. Li, M. Zhang, J. R. Martins, and C. Shu, “Efficient aerodynamic shape optimization with deep- learning-based geometric filtering,” AIAA Journal , pp. 1–17, 2020

    work page 2020

  2. [2]

    Hull-form stochastic optimization via computational-cost reduction methods,

    A. Serani, F. Stern, E. F. Campana, and M. Diez, “Hull-form stochastic optimization via computational-cost reduction methods,” Engineering with Computers , no. 2, 2021

    work page 2021

  3. [3]

    Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retraining

    A. Tripp, E. Daxberger, and J. M. Hern´ andez-Lobato, “Sample-efficient optimization in the latent space of deep generative models via weighted retraining,” arXiv preprint arXiv:2006.09191 , 2020

    work page Pith review arXiv 2006

  4. [4]

    Airfoil design parameterization and optimization using b \’ezier generative adversarial networks,

    K. Chiu, M. Fuge et al., “Airfoil design parameterization and optimization using b \’ezier generative adversarial networks,” arXiv preprint arXiv:2006.12496 , 2020

    arXiv 2006

  5. [5]

    A b-spline-based generative adversarial network model for fast interactive airfoil aerodynamic optimization,

    X. Du, P. He, and J. Martins, “A b-spline-based generative adversarial network model for fast interactive airfoil aerodynamic optimization,” in AIAA Scitech 2020 Forum , 2020, p. 2128

    work page 2020

  6. [6]

    Grammar variational autoencoder,

    M. J. Kusner, B. Paige, and J. M. Hern´ andez-Lobato, “Grammar variational autoencoder,” in International Conference on Machine Learning . PMLR, 2017, pp. 1945–1954

    work page 2017

  7. [7]

    Constrained Bayesian Optimization for Automatic Chemical Design

    R. R. Griffiths, “Constrained bayesian optimization for automatic chemical design,” arXiv preprint arXiv:1709.05501, 2017

    work page Pith review arXiv 2017

  8. [8]

    Aerodynamic design optimization and shape exploration using generative adversarial networks,

    W. Chen, K. Chiu, and M. Fuge, “Aerodynamic design optimization and shape exploration using generative adversarial networks,” in AIAA Scitech 2019 Forum , 2019, p. 2351

    work page 2019

Show all 60 references
  1. [9]

    Structured variationally auto-encoded optimiza- tion,

    X. Lu, J. Gonzalez, Z. Dai, and N. D. Lawrence, “Structured variationally auto-encoded optimiza- tion,” in International conference on machine learning . PMLR, 2018, pp. 3267–3275

    2018

  2. [10]

    Synthesizing the preferred inputs for neurons in neural networks via deep generator networks,

    A. Nguyen, A. Dosovitskiy, J. Yosinski, T. Brox, and J. Clune, “Synthesizing the preferred inputs for neurons in neural networks via deep generator networks,”Advances in neural information processing systems, vol. 29, pp. 3387–3395, 2016

    2016

  3. [11]

    Neural architecture optimization,

    R. Luo, F. Tian, T. Qin, E. Chen, and T.-Y. Liu, “Neural architecture optimization,” arXiv preprint arXiv:1808.07233, 2018

    2018 arXiv

  4. [12]

    Microstructural materials design via deep adversarial learning methodology,

    Z. Yang, X. Li, L. Catherine Brinson, A. N. Choudhary, W. Chen, and A. Agrawal, “Microstructural materials design via deep adversarial learning methodology,”Journal of Mechanical Design, vol. 140, no. 11, 2018

    2018

  5. [13]

    Automatic chemical design using a data-driven continuous representation of molecules,

    R. G´ omez-Bombarelli, J. N. Wei, D. Duvenaud, J. M. Hern´ andez-Lobato, B. S´ anchez-Lengeling, D. Sheberla, J. Aguilera-Iparraguirre, T. D. Hirzel, R. P. Adams, and A. Aspuru-Guzik, “Automatic chemical design using a data-driven continuous representation of molecules,” ACS c...

    2018

  6. [14]

    Survey of modeling and optimization strategies to solve high-dimensional design problems with computationally-expensive black-box functions,

    S. Shan and G. G. Wang, “Survey of modeling and optimization strategies to solve high-dimensional design problems with computationally-expensive black-box functions,” Structural and multidisci- plinary optimization , vol. 41, no. 2, pp. 219–241, 2010

    2010

  7. [15]

    Insights on transfer optimization: Because experience is the best teacher,

    A. Gupta, Y.-S. Ong, and L. Feng, “Insights on transfer optimization: Because experience is the best teacher,” IEEE Transactions on Emerging Topics in Computational Intelligence , vol. 2, no. 1, pp. 51–64, 2017

    2017

  8. [16]

    Curbing negative influences online for seamless transfer evolu- tionary optimization,

    B. Da, A. Gupta, and Y.-S. Ong, “Curbing negative influences online for seamless transfer evolu- tionary optimization,” IEEE transactions on cybernetics , vol. 49, no. 12, pp. 4365–4378, 2018

    2018

  9. [17]

    Toward adaptive knowledge transfer in multifactorial evolutionary computation,

    L. Zhou, L. Feng, K. C. Tan, J. Zhong, Z. Zhu, K. Liu, and C. Chen, “Toward adaptive knowledge transfer in multifactorial evolutionary computation,” IEEE Transactions on Cybernetics , 2020

    2020

  10. [18]

    Evolutionary multitasking for multiobjective optimization with subspace alignment and adaptive differential evolution,

    Z. Liang, H. Dong, C. Liu, W. Liang, and Z. Zhu, “Evolutionary multitasking for multiobjective optimization with subspace alignment and adaptive differential evolution,” IEEE Transactions on Cybernetics, 2020

    2020

  11. [19]

    A meta-knowledge transfer-based differential evolution for multitask optimization,

    J.-Y. Li, Z.-H. Zhan, K. C. Tan, and J. Zhang, “A meta-knowledge transfer-based differential evolution for multitask optimization,” IEEE Transactions on Evolutionary Computation , 2021

    2021

  12. [20]

    Evolutionary multitasking across single and multi- objective formulations for improved problem solving,

    B. Da, A. Gupta, Y. S. Ong, and F. Liang, “Evolutionary multitasking across single and multi- objective formulations for improved problem solving,” in Evolutionary Computation , 2016

    2016

  13. [21]

    A study on multiform multi-objective evolutionary optimization,

    L. Zhang, Y. Xie, J. Chen, L. Feng, C. Chen, and K. Liu, “A study on multiform multi-objective evolutionary optimization,” Memetic Computing , pp. 1–12, 2021

    2021

  14. [22]

    Multimodal optimization enhanced cooperative coevolution for large-scale optimization,

    X. Peng, Y. Jin, and H. Wang, “Multimodal optimization enhanced cooperative coevolution for large-scale optimization,” IEEE transactions on cybernetics , vol. 49, no. 9, pp. 3507–3520, 2018

    2018

  15. [23]

    Efficient generalized surrogate-assisted evolutionary algorithm for high- dimensional expensive problems,

    X. Cai, L. Gao, and X. Li, “Efficient generalized surrogate-assisted evolutionary algorithm for high- dimensional expensive problems,” IEEE Transactions on Evolutionary Computation , vol. 24, no. 2, pp. 365–379, 2019

    2019

  16. [24]

    A social learning particle swarm optimization algorithm for scalable opti- mization,

    R. Cheng and Y. Jin, “A social learning particle swarm optimization algorithm for scalable opti- mization,” Information Sciences, vol. 291, pp. 43–60, 2015

    2015

  17. [25]

    Committee-based active learning for surrogate-assisted particle swarm optimization of expensive problems,

    H. Wang, Y. Jin, and J. Doherty, “Committee-based active learning for surrogate-assisted particle swarm optimization of expensive problems,” IEEE transactions on cybernetics , vol. 47, no. 9, pp. 2664–2677, 2017

    2017

  18. [26]

    A tutorial on bayesian optimization,

    P. I. Frazier, “A tutorial on bayesian optimization,” arXiv preprint arXiv:1807.02811 , 2018

    2018 arXiv

  19. [27]

    Funneled bayesian optimization for design, tuning and control of autonomous systems,

    Martinez-Cantin and Ruben, “Funneled bayesian optimization for design, tuning and control of autonomous systems,” IEEE Transactions on Cybernetics , vol. 49, no. 4, pp. 1489–1500, 2019

    2019

  20. [28]

    Good practices for bayesian optimization of high dimensional structured spaces,

    E. Siivola, J. Gonzalez, A. Paleyes, and A. Vehtari, “Good practices for bayesian optimization of high dimensional structured spaces,” arXiv preprint arXiv:2012.15471 , 2020

    2012 arXiv

  21. [29]

    Calibrated and recalibrated expected improvements for bayesian optimization,

    Z. Guo, Y. Ong, and H. Liu, “Calibrated and recalibrated expected improvements for bayesian optimization,” Structural and Multidisciplinary Optimization , 2021

    2021

  22. [30]

    Generalizing transfer bayesian optimization to source-target heterogeneity,

    A. Tan, A. Gupta, and Y. S. Ong, “Generalizing transfer bayesian optimization to source-target heterogeneity,” IEEE Transactions on Automation Science and Engineering , vol. PP, no. 99, 2020

    2020

  23. [31]

    Auto-encoding variational bayes,

    D. P. Kingma and M. Welling, “Auto-encoding variational bayes,” arXiv preprint arXiv:1312.6114 , 2013

    2013 arXiv

  24. [32]

    Generative adversarial nets,

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets,” in Advances in neural information processing systems , 2014, pp. 2672–2680

    2014

  25. [33]

    Generative adversarial networks: An overview,

    A. Creswell, T. White, V. Dumoulin, K. Arulkumaran, B. Sengupta, and A. A. Bharath, “Generative adversarial networks: An overview,” IEEE Signal Processing Magazine , vol. 35, no. 1, pp. 53–65, 2018. 21

    2018

  26. [34]

    Knowledge transfer through machine learning in aircraft design,

    A. T. W. Min, R. Sagarna, A. Gupta, Y.-S. Ong, and C. K. Goh, “Knowledge transfer through machine learning in aircraft design,” IEEE Computational Intelligence Magazine , vol. 12, no. 4, pp. 48–60, 2017

    2017

  27. [35]

    Efficient transfer learning method for automatic hyperparameter tuning,

    D. Yogatama and G. Mann, “Efficient transfer learning method for automatic hyperparameter tuning,” in Artificial intelligence and statistics , 2014, pp. 1077–1085

    2014

  28. [36]

    Evolutionary transfer optimization-a new frontier in evolutionary computation research,

    K. C. Tan, L. Feng, and M. Jiang, “Evolutionary transfer optimization-a new frontier in evolutionary computation research,” IEEE Computational Intelligence Magazine , vol. 16, no. 1, pp. 22–33, 2021

    2021

  29. [37]

    A review on evolutionary multi-task optimization: Trends and challenges,

    T. Wei, S. Wang, J. Zhong, D. Liu, and J. Zhang, “A review on evolutionary multi-task optimization: Trends and challenges,” IEEE Transactions on Evolutionary Computation , 2021

    2021

  30. [38]

    A flexible transfer learning framework for bayesian optimization with convergence guarantee,

    T. n. T. Joy, S. Rana, S. Gupta, and S. Venkatesh, “A flexible transfer learning framework for bayesian optimization with convergence guarantee,” Expert Systems with Applications , vol. 115, pp. 656–672, 2019

    2019

  31. [39]

    Multi-objective multi-tasking optimization based on incremental learning,

    J. Lin, H.-L. Liu, B. Xue, M. Zhang, and F. Gu, “Multi-objective multi-tasking optimization based on incremental learning,” IEEE Transactions on Evolutionary Computation , 2019

    2019

  32. [40]

    Multi-task bayesian optimization,

    K. Swersky, J. Snoek, and R. P. Adams, “Multi-task bayesian optimization,” in Advances in neural information processing systems, 2013, pp. 2004–2012

    2013

  33. [41]

    Reducing local optima in single-objective problems by multi-objectivization,

    J. D. Knowles, R. A. Watson, and D. W. Corne, “Reducing local optima in single-objective problems by multi-objectivization,” in International Conference on Evolutionary Multi-criterion Optimiza- tion, 2001

    2001

  34. [42]

    Multiobjectivization by decomposition of scalar cost func- tions,

    J. Handl, S. C. Lovell, and J. Knowles, “Multiobjectivization by decomposition of scalar cost func- tions,” in Parallel Problem Solving from Nature-ppsn X, International Conference Dortmund, Ger- many, September, 2008

    2008

  35. [43]

    Parallel multi-fidelity expected improvement method for efficient global optimization,

    Z. Guo, Q. Wang, L. Song, and J. Li, “Parallel multi-fidelity expected improvement method for efficient global optimization,” Structural and Multidisciplinary Optimization , 2021

    2021

  36. [44]

    Analysis of dataset selection for multi-fidelity surrogates for a turbine problem,

    Z. Guo, L. Song, C. Park, J. Li, and R. T. Haftka, “Analysis of dataset selection for multi-fidelity surrogates for a turbine problem,” Structural and Multidisciplinary Optimization , vol. 57, no. 6, pp. 2127–2142, 2018

    2018

  37. [45]

    Multi-fidelity bayesian optimization via deep neural net- works,

    S. Li, W. Xing, R. Kirby, and S. Zhe, “Multi-fidelity bayesian optimization via deep neural net- works,” Advances in Neural Information Processing Systems , vol. 33, 2020

    2020

  38. [46]

    Theoretical and numerical constraint-handling techniques used with evolutionary algo- rithms: a survey of the state of the art,

    C. Coello, “Theoretical and numerical constraint-handling techniques used with evolutionary algo- rithms: a survey of the state of the art,” Computer Methods in Applied Mechanics and Engineering , vol. 191, no. 11–12, pp. 1245–1287, 2002

    2002

  39. [47]

    Nash game based efficient global optimization for large-scale design problems,

    S. Xu and H. Chen, “Nash game based efficient global optimization for large-scale design problems,” Journal of Global Optimization , vol. 71, 2018

    2018

  40. [48]

    Taking the human out of the loop: A review of bayesian optimization,

    B. Shahriari, K. Swersky, Z. Wang, R. P. Adams, and N. De Freitas, “Taking the human out of the loop: A review of bayesian optimization,” Proceedings of the IEEE , vol. 104, no. 1, pp. 148–175, 2015

    2015

  41. [49]

    Gaussian processes for machine learning (gpml) toolbox,

    C. E. Rasmussen and H. Nickisch, “Gaussian processes for machine learning (gpml) toolbox,” The Journal of Machine Learning Research , vol. 11, pp. 3011–3015, 2010

    2010

  42. [50]

    Parallel gaussian process optimization with upper confidence bound and pure exploration,

    E. Contal, D. Buffoni, A. Robicquet, and N. Vayatis, “Parallel gaussian process optimization with upper confidence bound and pure exploration,” in Joint European Conference on Machine Learning and Knowledge Discovery in Databases , 2013, pp. 225–240

    2013

  43. [51]

    Efficient global optimization of expensive black-box functions,

    D. R. Jones, M. Schonlau, and W. J. Welch, “Efficient global optimization of expensive black-box functions,” Journal of Global optimization , vol. 13, no. 4, pp. 455–492, 1998

    1998

  44. [52]

    Infogan: Interpretable representation learning by information maximizing generative adversarial nets,

    X. Chen, Y. Duan, R. Houthooft, J. Schulman, I. Sutskever, and P. Abbeel, “Infogan: Interpretable representation learning by information maximizing generative adversarial nets,” in Advances in neural information processing systems , 2016, pp. 2172–2180. 22

    2016

  45. [53]

    Metric-based mathematical derivation of efficient airfoil design variables,

    D. J. Poole, C. B. Allen, and T. Rendall, “Metric-based mathematical derivation of efficient airfoil design variables,” AIAA Journal , vol. 53, no. 5, pp. 1349–1361, 2015

    2015

  46. [54]

    Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (cma-es),

    N. Hansen, S. D. M¨ uller, and P. Koumoutsakos, “Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (cma-es),” Evolutionary Computation , 2003

    2003

  47. [55]

    A fast kriging-assisted evolutionary algorithm based on incremental learn- ing,

    D. Zhan and H. Xing, “A fast kriging-assisted evolutionary algorithm based on incremental learn- ing,” IEEE Transactions on Evolutionary Computation , vol. PP, no. 99, pp. 1–1, 2021

    2021

  48. [56]

    Large sample properties of simulations using latin hypercube sampling,

    M. Stein, “Large sample properties of simulations using latin hypercube sampling,” Technometrics, vol. 29, no. 2, pp. 143–151, 1987

    1987

  49. [57]

    Hierarchical surrogate-assisted evolutionary multi-scenario airfoil shape optimization,

    H. Wang, J. Doherty, and Y. Jin, “Hierarchical surrogate-assisted evolutionary multi-scenario airfoil shape optimization,” in 2018 IEEE Congress on Evolutionary Computation (CEC) . IEEE, 2018, pp. 1–8

    2018

  50. [58]

    Summary of low-speed airfoil data, vol. 5,

    G. A. Williamson, B. D. McGranahan, B. A. Broughton, R. W. Deters, J. B. Brandt, and M. S. Selig, “Summary of low-speed airfoil data, vol. 5,” University of Illinois, Champaign, IL , vol. 204, 2012

    2012

  51. [59]

    Xfoil: An analysis and design system for low reynolds number airfoils,

    M. Drela, “Xfoil: An analysis and design system for low reynolds number airfoils,” in Low Reynolds number aerodynamics. Springer, 1989, pp. 1–12

    1989

  52. [60]

    Generative design of decorative architectural parts,

    Y. Zhang, C. O. Chan, J. Zheng, S. T. Lie, and Z. Guo, “Generative design of decorative architectural parts,” The Visual Computer , no. 1, pp. 1–17, 2021. 23

    2021

Pith tools

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