Pith. sign in

REVIEW 4 major objections 6 minor 48 references

One learned model maps any constraint threshold to a near-optimal design, so changing feasibility limits no longer means restarting expensive optimization.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-30 21:53 UTC pith:TYF65ETQ

load-bearing objection Useful amortization idea for multi-threshold ECOPs, but the N-scaling win rates partly rest on baselines that are forbidden from pooling the shared (f,c) data. the 4 major comments →

arxiv 2607.23448 v1 pith:TYF65ETQ submitted 2026-07-26 cs.NE cs.AIcs.LG

Constraint-Bound Agnostic Bayesian Optimization: One Model for All Thresholds

Jin Wang , Xi Lin , Handing Wang This is my paper
classification cs.NE cs.AIcs.LG
keywords constrained Bayesian optimizationparametric optimizationthreshold-to-solution mappingexpensive black-box constraintsGaussian process surrogatesintent-guided recommendationengineering design
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Engineers often cannot fix constraint thresholds in advance and must try many feasibility settings on the same expensive black-box design problem. Standard constrained Bayesian optimization treats every new threshold vector as a fresh problem, wasting evaluations and ignoring that the underlying objective and constraints are shared. This paper claims that a neural network can be trained, inside a single Bayesian optimization run, to map any threshold vector directly to a high-quality solution. After roughly two hundred shared evaluations the model answers arbitrary new threshold queries instantly; one local Bayesian refinement step then corrects residual errors. The practical payoff is amortization: the more thresholds an engineer later queries, the larger the advantage over methods that re-solve each setting from scratch, plus a simple intent-guided recommender that suggests better thresholds matching stated tighten/lock/loosen preferences.

Core claim

After a single shared Bayesian-optimization budget, a parametric constraint model learns a reusable map from continuous constraint-threshold vectors to near-optimal designs. Direct prediction plus one local LogcEI refinement then yields better feasible solutions than independent constrained BO or evolutionary runs on the same total evaluation budget, and the advantage grows as the number of queried thresholds increases.

What carries the argument

The parametric constraint model (PCM): an MLP h_φ(θ) that outputs a candidate design for any threshold vector θ. It is trained by gradient descent on a feasibility-aware surrogate objective built from GP lower/upper confidence bounds, optionally regularized for monotonicity under threshold relaxation, then refined by one local constrained expected-improvement step.

Load-bearing premise

Independent Gaussian-process surrogates supply gradients accurate enough, from only about two hundred expensive evaluations, to train one global neural map over the whole continuous family of thresholds.

What would settle it

On a problem with a highly irregular or multi-modal feasible region, train CBA-BO once with the stated budget, then query one thousand random unseen thresholds; if direct prediction plus one-step refinement loses to per-threshold LogcEI given the same total evaluations, the amortization claim fails.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Engineers can explore many feasibility–performance trade-offs after one optimization campaign instead of restarting for every new threshold.
  • Win-rate gains over independent constrained BO widen as the number of queried thresholds grows, because learning cost is amortized.
  • Direct PCM prediction already recovers high feasibility on many problems; one local refinement step lifts feasibility above 90–99 % on the harder cases tested.
  • An intent-guided recommender can propose new thresholds that improve the objective while obeying user tighten/lock/loosen preferences without manual trial-and-error.

Where Pith is reading between the lines

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

  • The same threshold-to-solution map could serve as a warm-start generator for multi-fidelity or multi-objective constrained design loops that also vary resource or preference parameters.
  • If the PCM’s monotonicity regularizer is strengthened, the model may become a differentiable surrogate for sensitivity analysis of active constraints across the threshold domain.
  • Failure modes on narrow or disconnected feasible sets suggest a natural hybrid: fall back to local trust-region BO automatically when PCM prediction variance or violation exceeds a threshold.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes CBA-BO, a framework for expensive constrained optimization in which the constraint thresholds θ vary over a domain Θ. A parametric constraint model (PCM), an MLP h_φ(θ), is trained concurrently with Bayesian optimization using gradients of GP-based LCB/UCB surrogate objectives (Eqs. 9–13), plus a monotonicity regularizer (Eq. 14) encoding that relaxing thresholds cannot increase the optimal objective. After a single 200-evaluation training run, the PCM predicts solutions for arbitrary unseen thresholds; a one-step local LogcEI refinement (Eq. 17) corrects prediction errors. An intent-guided threshold recommendation utility (Eqs. 18–22) is demonstrated on the car side-impact problem. Experiments on 16 problems compare against CMA-ES, cEI, ALBO, SCBO, and LogcEI under a protocol where N thresholds share a fixed total budget; win rates grow with N (Figure 2, Tables 12–13), and feasibility at N=1000 is reported with and without refinement (Table 1).

Significance. If the amortization claim holds under fair controls, this is a practically useful contribution: engineers genuinely re-solve ECOPs under varying thresholds, and a single 200-evaluation training run yielding a reusable threshold→solution map with millisecond inference would be valuable. The empirical work is substantial: 16 problems, 10 runs, win-rate protocol against five baselines, N up to 1000, ablations of PCM guidance and monotonic regularization, a PCM-vs-random-initialization refinement control (Table 6), pairwise monotonicity tests (Table 7), hyperparameter sensitivity (Tables 9–11), and honest reporting that direct prediction is unreliable on problems with active or switching constraints (e.g., 39.5% feasibility on g01, Table 1). The acknowledgment that the method degrades for high-D or irregular feasible regions is appropriately candid.

major comments (4)
  1. [§Experimental Settings, Figure 2, Tables 12–13] The headline N-scaling claim rests on a protocol that forbids baselines from pooling data across thresholds, even though f and c_j are the *same* functions for every threshold — only the bound θ_j changes, so every evaluation (x, f(x), c(x)) is valid training data for all N ECOPs. CBA-BO itself exploits exactly this (its GPs are fit on the pooled database D_t). A pooled-GP LogcEI baseline — one ModelListGP fit on all evaluations, with the acquisition re-optimized per threshold at (170/N)+2 evaluations each — would match CBA-BO's budget and data access with no learned PCM. If that baseline is competitive, the Figure 2 trend reflects trivial data sharing rather than the PCM, which is the claimed contribution. This control is load-bearing and must be added.
  2. [§Experimental Settings] Related but distinct: at N=50 each baseline gets (170/50)+2 ≈ 5.4 evaluations per threshold after the 30 shared LHS points — below what any GP-based method needs to fit even a single-output surrogate meaningfully. The monotone increase of win rate with N therefore partly reflects driving baselines into a degenerate regime rather than an intrinsic amortization advantage. The paper should either (a) report absolute objective values as a function of N for all methods (Tables 14–18 show only one θ per N), or (b) add a fixed per-threshold budget comparison with total-cost accounting, so the efficiency claim is separated from the budget-splitting artifact.
  3. [Abstract, Table 1, §Local Refinement] The abstract's claim that CBA-BO 'directly predicts solutions for arbitrary unseen threshold configurations without additional optimization' is in tension with Table 1: direct prediction feasibility is 39.5% (g01), 68.2% (ThreeTruss), 69.2% (g07), 72.1% (Car), and KeaneBump is not improved at all by refinement (92.5±15.8% both rows). The practical method is PCM + per-threshold local LogcEI refinement, which is itself a small per-threshold BO. The framing 'one model for all thresholds' should be qualified, and the conditions under which direct prediction suffices (inactive constraints, smooth solution manifolds, per the Appendix visualization) should be stated in the main text, not only the Appendix.
  4. [§Analysis of PCM-based Initialization, Table 6] The control meant to isolate the PCM's contribution trains GPs on 200 *randomly sampled* points, whereas CBA-BO trains on the BO-collected database D_T. This conflates two things: PCM-guided initialization and PCM-guided data acquisition (the batch strategy of Eq. 16 selects candidates that shape D_T). The clean control is LogcEI refinement initialized from a random point but with GPs fit on CBA-BO's own D_T (or on data collected by threshold-agnostic BO). Two rows (g01: 47.5%, GKXWC2: 45.5%) already show the random-init baseline winning; the authors' explanation (simple landscapes) is plausible but the confound should be removed before concluding the PCM learning strategy is what helps.
minor comments (6)
  1. [Figure 2] Presenting win rates as a dense grid of numbers (16 problems × 5 N values × 5 baselines) is hard to parse; a line plot of win rate vs. N per problem (small multiples) with baseline panels would communicate the central trend far better.
  2. [§Experimental Results and Analysis] The main text underplays the N=1 column of Table 13: against LogcEI, CBA-BO scores 0% on g01, GKXWC1, KeaneBump, PressureVessel and 10% on g07. Since N=1 is the only setting where the baseline comparison is not affected by budget splitting, these losses deserve explicit discussion rather than the single sentence about LogcEI being 'highly competitive'.
  3. [Eq. (11)] The surrogate objective switches discontinuously between LCB_f and total violation at the feasibility boundary; the text says gradients come from 'GP surrogate gradients' but does not discuss how the max(0,·) kinks in Eq. (10) and the switch in Eq. (11) are handled in the chain rule of Eq. (13). A sentence on subgradient choices would help reproducibility.
  4. [Eq. (14)] The monotonicity regularizer is applied to the LCB surrogate f̂, not the true f; since the monotonicity property holds for f* but only approximately for LCB_f (which includes a −βσ term), the regularizer can penalize legitimate uncertainty-driven variation. Please comment on this choice, and on the g07 regression in Table 7 (81.9% → 72.5%).
  5. [Table 4] The threshold domains (e.g., [210×3, 110×6] for g01) are given without justification. How were θ_U values chosen relative to the constraint values at the unconstrained optimum? The difficulty of the learned mapping depends strongly on this choice.
  6. [General] Code/data availability is not stated; given the number of tuned components (Table 5 lists ~20 settings), release of implementation would substantially strengthen the paper. Typos: 'A extra analysis' (§Ablation Study); inconsistent spacing artifacts throughout ('optimizationproblems', 'constraint-bounds in practice are hard to be determined').

Circularity Check

0 steps flagged

No circularity: empirical amortization claim evaluated on held-out black-box thresholds against external baselines, not forced by construction.

full rationale

CBA-BO trains a parametric map h_φ(θ)≈x*(θ) by minimizing a GP-surrogate feasibility-aware loss (Eqs. 9–13) over sampled thresholds, then reports true black-box objective/feasibility on held-out θ queries and win rates versus independent constrained-BO and evolutionary baselines (Figure 2, Table 1). Direct prediction is not tautological: feasibility on N=1000 can be as low as ~39.5% (g01) before refinement, so success is not guaranteed by the training identity. Monotonic regularization is a soft additive prior (α r_mono), not a definition that forces reported objectives. Related-work citations to Pareto-set learning motivate the architecture but do not supply a uniqueness theorem or fitted identity that the central N-scaling claim reduces to. Experimental-protocol concerns (whether baselines should pool evaluations) are fairness/validity issues, not circular derivation. The paper is self-contained against external benchmarks; no step reduces a claimed prediction to its inputs by construction.

Axiom & Free-Parameter Ledger

5 free parameters · 4 axioms · 3 invented entities

The central amortization claim rests on standard BO/GP machinery, a domain monotonicity prior, and several hand-set training and architecture choices that define the PCM learning dynamics. No new physical entities; the ‘invented’ pieces are methodological constructs (PCM, CBA-BO loop, recommendation utility) whose only evidence is the paper’s own benchmarks.

free parameters (5)
  • LCB/UCB coefficient β = 0.5
    Fixed at 0.5 for objective LCB and constraint UCB guidance during PCM training; controls exploration in surrogate gradients and is validated only by limited sensitivity wins.
  • Monotonic regularization weight α = 0.1
    Added in second half of training to penalize positive ∂f̂/∂θ sensitivity; default 0.1 chosen by authors.
  • PCM architecture and training schedule = 3×256 MLP; 50 steps/iter; 200 evals
    MLP width/depth (3×256), Adam lr 1e-3→1e-5, 50 inner steps, batch 64, 34 BO iterations × batch 5 after 30 LHS—collectively determine what map is learned under the 200-eval budget.
  • Local refinement radius r = 0.25
    One-step LogcEI neighborhood size in normalized design space; directly affects whether refinement can rescue bad PCM predictions.
  • Recommendation utility weights (w_f, λ_h, λ_r, ε_j) = w_f=100, λ_h=10000, λ_r=50
    Hand-set scores that pick θ_rec from local candidates; case-study outcomes depend on these preferences-as-numbers.
axioms (4)
  • domain assumption Objective and constraints are well-modeled by independent GPs with stationary kernels sufficient for gradient-based PCM updates from limited data.
    Invoked throughout GP-guided learning and local refinement; Conclusions admit degradation in high-D or irregular regions.
  • domain assumption Relaxing thresholds cannot worsen the true optimal objective (monotonicity of f* with respect to θ componentwise relaxation).
    Used to justify r_mono in Eqs. 14–15; true for nested feasible sets on minimization problems if the same global optimum notion is used.
  • standard math Standard constrained BO acquisition theory (EI/logcEI, LCB/UCB bandits) is a valid sample-efficient backbone for ECOPs.
    Background from cited BO literature; not re-proved.
  • ad hoc to paper A single MLP can approximate the set-valued or switching active-constraint solution map x*(θ) well enough over the chosen Θ boxes.
    Implicit in PCM design; visualizations show harder cases (ThreeTruss) where active sets change and prediction errors rise.
invented entities (3)
  • Parametric Constraint Model (PCM) h_φ(θ) no independent evidence
    purpose: Neural map from threshold vectors to predicted optimal designs enabling amortization across θ.
    Core proposed object; evidence is internal benchmark performance, not an external independently measured phenomenon.
  • CBA-BO training loop (GP-guided PCM + batch sampling + one-step refinement) no independent evidence
    purpose: Unified algorithm that learns and exploits the threshold–solution relationship under an expensive evaluation budget.
    Methodological bundle evaluated only in this paper’s protocol.
  • Intent-guided constraint-bound recommendation utility S(θ) no independent evidence
    purpose: Select new thresholds from tighten/lock/loosen intents using PCM predictions and GP feasibility filters.
    Application layer on top of PCM; demonstrated in one Car case study.

pith-pipeline@v1.2.0-grok45-kimik3 · 34686 in / 3898 out tokens · 77740 ms · 2026-07-30T21:53:28.147601+00:00 · methodology

0 comments
read the original abstract

Expensive constrained optimization problems in real-world industry design often involve constraint thresholds that are difficult to determine in advance. Engineers may need to adjust constraint thresholds to explore different feasibility-performance trade-offs, requiring solutions under a wide range of threshold settings. However, existing constrained Bayesian optimization methods treat each threshold configuration independently, leading to repeated optimization and failing to exploit the shared relationship among continuously varying thresholds. To address this challenge, we propose constraint-bound agnostic Bayesian optimization (CBA-BO), a learning-based framework that learns a parametric constraint model mapping thresholds to optimal solutions. Once learned, CBA-BO directly predicts solutions for arbitrary unseen threshold configurations without additional optimization, with a one-step Bayesian optimization refinement further improving solution quality. Experiments on benchmark and engineering problems demonstrate that CBA-BO learns a transferable threshold-solution mapping, enabling efficient prediction and optimization for arbitrary threshold queries. An intent-guided constraint-bound recommendation mechanism is further developed to improve objective performance while satisfying user-specified constraint preferences.

Figures

Figures reproduced from arXiv: 2607.23448 by Handing Wang, Jin Wang, Xi Lin.

Figure 1
Figure 1. Figure 1: Architecture comparison of three paradigms for [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Win rate (%) of CBA-BO with one-step refinement against baseline methods under different numbers of threshold [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of PCM-predicted solutions and grid-search optima on three 2D problems. For each problem, the left [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Visualization of PCM prediction and LogcEI refinement on three 2D problems. Gold stars denote grid-search optima, [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗

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

48 extracted references · 5 linked inside Pith

  1. [1]

    Communication, Simulation, and Intelligent Agents: Implications of Personal Intelligent Machines for Medical Education

    Clancey, William J. Communication, Simulation, and Intelligent Agents: Implications of Personal Intelligent Machines for Medical Education. Proceedings of the Eighth International Joint Conference on Artificial Intelligence (IJCAI-83)

  2. [2]

    Classification Problem Solving

    Clancey, William J. Classification Problem Solving. Proceedings of the Fourth National Conference on Artificial Intelligence

  3. [3]

    , title =

    Robinson, Arthur L. , title =. 1980 , doi =. https://science.sciencemag.org/content/208/4447/1019.full.pdf , journal =

    1980

  4. [4]

    New Ways to Make Microcircuits Smaller---Duplicate Entry

    Robinson, Arthur L. New Ways to Make Microcircuits Smaller---Duplicate Entry. Science

  5. [5]

    Clancey and Glenn Rennels , abstract =

    Diane Warner Hasling and William J. Clancey and Glenn Rennels , abstract =. Strategic explanations for a diagnostic consultation system , journal =. 1984 , issn =. doi:https://doi.org/10.1016/S0020-7373(84)80003-6 , url =

    doi:10.1016/s0020-7373(84)80003-6 1984

  6. [6]

    and Rennels, Glenn R

    Hasling, Diane Warner and Clancey, William J. and Rennels, Glenn R. and Test, Thomas. Strategic Explanations in Consultation---Duplicate. The International Journal of Man-Machine Studies

  7. [7]

    Poligon: A System for Parallel Problem Solving

    Rice, James. Poligon: A System for Parallel Problem Solving

  8. [8]

    Transfer of Rule-Based Expertise through a Tutorial Dialogue

    Clancey, William J. Transfer of Rule-Based Expertise through a Tutorial Dialogue

  9. [9]

    The Engineering of Qualitative Models

    Clancey, William J. The Engineering of Qualitative Models

  10. [10]

    2017 , eprint=

    Attention Is All You Need , author=. 2017 , eprint=

    2017

  11. [11]

    Pluto: The 'Other' Red Planet

    NASA. Pluto: The 'Other' Red Planet

  12. [12]

    and de Freitas, Nando , journal=

    Shahriari, Bobak and Swersky, Kevin and Wang, Ziyu and Adams, Ryan P. and de Freitas, Nando , journal=. Taking the Human Out of the Loop: A Review of Bayesian Optimization , year=

  13. [13]

    International conference on artificial intelligence and statistics , pages=

    Scalable constrained Bayesian optimization , author=. International conference on artificial intelligence and statistics , pages=. 2021 , organization=

    2021

  14. [14]

    Recent advances in optimization and modeling of contemporary problems , pages=

    Bayesian optimization , author=. Recent advances in optimization and modeling of contemporary problems , pages=. 2018 , publisher=

    2018

  15. [15]

    , author=

    Bayesian optimization with inequality constraints. , author=. ICML , volume=

  16. [16]

    Bayesian Analysis , number =

    Benjamin Letham and Brian Karrer and Guilherme Ottoni and Eytan Bakshy , title =. Bayesian Analysis , number =

  17. [17]

    Advances in neural information processing systems , volume=

    Multi-task bayesian optimization , author=. Advances in neural information processing systems , volume=

  18. [18]

    Advances in neural information processing systems , volume=

    Multi-information source optimization , author=. Advances in neural information processing systems , volume=

  19. [19]

    Advances in neural information processing systems , volume=

    Pareto set learning for expensive multi-objective optimization , author=. Advances in neural information processing systems , volume=

  20. [20]

    arXiv preprint arXiv:2505.04037 , year=

    Learning based convex approximation for constrained parametric optimization , author=. arXiv preprint arXiv:2505.04037 , year=

    Pith/arXiv arXiv

  21. [21]

    arXiv preprint arXiv:2512.20270 , year=

    Optimality-Informed Neural Networks for Solving Parametric Optimization Problems , author=. arXiv preprint arXiv:2512.20270 , year=

    arXiv

  22. [22]

    IEEE Communications Letters , volume=

    Unsupervised learning for parametric optimization , author=. IEEE Communications Letters , volume=. 2020 , publisher=

    2020

  23. [23]

    A Distributed Surrogate-Assisted Evolutionary Algorithm for Heterogeneously Expensive Constrained Optimization , year=

    Wang, Jin and Wang, Xilu and Wang, Handing and Jin, Yaochu , journal=. A Distributed Surrogate-Assisted Evolutionary Algorithm for Heterogeneously Expensive Constrained Optimization , year=

  24. [24]

    A Surrogate-Assisted Evolutionary Framework With Regions of Interests-Based Data Selection for Expensive Constrained Optimization , year=

    Song, Zhenshou and Wang, Handing and Jin, Yaochu , journal=. A Surrogate-Assisted Evolutionary Framework With Regions of Interests-Based Data Selection for Expensive Constrained Optimization , year=

  25. [25]

    Constrained Bayesian Optimization: A Review , year=

    Amini, Sasan and Vannieuwenhuyse, Inneke and Morales-Hernández, Alejandro , journal=. Constrained Bayesian Optimization: A Review , year=

  26. [26]

    arXiv preprint arXiv:1403.5607 , year=

    Bayesian optimization with unknown constraints , author=. arXiv preprint arXiv:1403.5607 , year=

    Pith/arXiv arXiv

  27. [27]

    The computer journal , volume=

    A simplex method for function minimization , author=. The computer journal , volume=. 1965 , publisher=

    1965

  28. [28]

    Advances in neural information processing systems , volume=

    Scalable global optimization via local Bayesian optimization , author=. Advances in neural information processing systems , volume=

  29. [29]

    Biometrika , volume=

    On the likelihood that one unknown probability exceeds another in view of the evidence of two samples , author=. Biometrika , volume=. 1933 , publisher=

    1933

  30. [30]

    Advances in neural information processing systems , volume=

    Unexpected improvements to expected improvement for bayesian optimization , author=. Advances in neural information processing systems , volume=

  31. [31]

    arXiv preprint arXiv:0912.3995 , year=

    Gaussian process optimization in the bandit setting: No regret and experimental design , author=. arXiv preprint arXiv:0912.3995 , year=

    Pith/arXiv arXiv

  32. [32]

    arXiv preprint arXiv:2004.12214 , year=

    Learning to guide random search , author=. arXiv preprint arXiv:2004.12214 , year=

    Pith/arXiv arXiv 2004

  33. [33]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    Deep residual learning for image recognition , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

  34. [34]

    Technometrics , volume=

    Modeling an augmented Lagrangian for blackbox constrained optimization , author=. Technometrics , volume=. 2016 , publisher=

    2016

  35. [35]

    Towards a new evolutionary computation: Advances in the estimation of distribution algorithms , pages=

    The CMA evolution strategy: a comparing review , author=. Towards a new evolutionary computation: Advances in the estimation of distribution algorithms , pages=. 2006 , publisher=

    2006

  36. [36]

    Proceedings of the 14th annual conference on Genetic and evolutionary computation , pages=

    A (1+ 1)-CMA-ES for constrained optimisation , author=. Proceedings of the 14th annual conference on Genetic and evolutionary computation , pages=

  37. [37]

    Journal of Applied Mechanics , volume=

    Problem definitions and evaluation criteria for the CEC 2006 special session on constrained real-parameter optimization , author=. Journal of Applied Mechanics , volume=

    2006

  38. [38]

    Nanyang Technological University, Singapore , volume=

    Problem definitions and evaluation criteria for the CEC 2010 competition on constrained real-parameter optimization , author=. Nanyang Technological University, Singapore , volume=

    2010

  39. [39]

    Yu et al

    Fast and accurate Bayesian optimization with pre-trained transformers for constrained engineering problems: R. Yu et al.. , author=. Structural and Multidisciplinary Optimization , volume=. 2025 , publisher=

    2025

  40. [40]

    Proceedings of the conference on adaptive computing in engineering design and control , volume=

    Experiences with optimizers in structural design , author=. Proceedings of the conference on adaptive computing in engineering design and control , volume=

  41. [41]

    Computers & Structures , volume=

    Mixed variable structural optimization using firefly algorithm , author=. Computers & Structures , volume=. 2011 , publisher=

    2011

  42. [42]

    Engineering computations , volume=

    Bat algorithm: a novel approach for global engineering optimization , author=. Engineering computations , volume=. 2012 , publisher=

    2012

  43. [43]

    Journal of Mechanical Design , volume=

    Nonlinear integer and discrete programming in mechanical design optimization , author=. Journal of Mechanical Design , volume=

  44. [44]

    International journal of vehicle design , volume=

    Optimisation and robustness for crashworthiness of side impact , author=. International journal of vehicle design , volume=. 2001 , publisher=

    2001

  45. [45]

    Advances in neural information processing systems , volume=

    BoTorch: A framework for efficient Monte-Carlo Bayesian optimization , author=. Advances in neural information processing systems , volume=

  46. [46]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Parametric Pareto set learning for expensive multi-objective optimization , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  47. [47]

    IEEE Transactions on Evolutionary Computation , year=

    Parametric Pareto Set Learning: Amortizing Multi-Objective Optimization With Parameters , author=. IEEE Transactions on Evolutionary Computation , year=

  48. [48]

    arXiv preprint arXiv:2203.15386 , year=

    Pareto set learning for neural multi-objective combinatorial optimization , author=. arXiv preprint arXiv:2203.15386 , year=

    Pith/arXiv arXiv