Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Monte Carlo Tree Search based Space Transfer for Black-box Optimization

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read MCTS-transfer grows a search-space tree from source tasks and adaptively rewires it during optimization, and the experiments report better rankings than other transfer methods on BBOB, real-world problems, Design-Bench, and HPOB.

desk verdict MCTS-transfer is a reasonable, well-tested extension of LA-MCTS to transfer BO, but the main empirical claim is undercut by tuning the similarity measure on the test benchmarks and a UCB formula inconsistency. read the letter →

arxiv 2412.07186 v1 pith:KHC5H7WD submitted 2024-12-10 cs.LG cs.AI

Shukuan Wang , Ke Xue , Lei Song , Xiaobin Huang , Chao Qian This is my paper · ORCID
classification cs.LGcs.AI
keywords searchspacetransferBayesianoptimizationMonteCarlotreeblack-boxlearningadaptivetasksimilarityhyperparameterpartition
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

Fresh Bayesian optimization runs are usually started from scratch even when related tasks have already been solved. This paper proposes MCTS-transfer, a method that transfers the search space itself from source tasks to a new target task by growing a Monte Carlo tree whose nodes are subspaces. The tree is first built from source-task data to warm-start the target, then rebuilt during optimization so that source tasks judged more similar to the target get larger weights in the node-potential calculation. The authors aim to show that this adaptive weighting lets the method keep the benefits of transfer even when some source tasks are dissimilar, and their experiments on synthetic functions, real-world problems, Design-Bench, and hyperparameter optimization report better rankings than existing search-space transfer methods. If correct, the method makes transferable Bayesian optimization more reliable when task similarity is unknown ahead of time.

What carries the argument

The central object is the MCTS search tree, where the root is the entire domain and each node is a subspace produced by k-means clustering of the samples inside its parent followed by a binary classifier that separates good from bad regions. Node value is the potential $p_m$ of Eq. (3): a decay-weighted blend of the average target-task value in the subspace and a similarity-weighted average of source-task values, and node selection follows the UCB score of Eq. (1), the empirical mean potential plus an exploration bonus that grows with the logarithm of the parent's visit count. The load-bearing update is the adaptive weight $w_i$ of Eq. (4), computed from the rank of each source task's distance to the target, which changes with every new target sample and feeds both the node potentials and the reconstruction decision.

What would settle it

Construct a synthetic family of target functions whose true similarity ordering is known in advance but where the best-$N$ sample-mean distance mis-ranks the sources, for example by shifting the optimum along a direction where the best samples of a dissimilar source are closer to the target's early samples than a truly similar source's samples. If MCTS-transfer then assigns higher weights to the misleading source and its tree reconstruction moves the chosen subspace away from the true optimum, the central adaptive-weighting claim is contradicted.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that search-space transfer for Bayesian optimization can be made adaptive by representing the partition of the search space as a Monte Carlo tree. Each node of the tree is a subspace; its potential is a weighted combination of the average objective values of the source-task samples and target-task samples inside that subspace, with source weights recomputed after every new target evaluation according to how close each source task is to the target (distance between means of the best few samples). Because the left child of every node is supposed to be the better subspace, a violation of that ordering triggers a tree-reconstruction step that prunes and regrows the subtree, letting the search space track the target task as observations accumulate. The paper reports that this two-stage procedure, an offline pre-learned tree for warm-start followed by online adaptive partition, outperforms the compared search-space transfer and transfer-BO baselines across BBOB, real-world, Design-Bench, and HPO-B benchmarks in both similar and mixed transfer settings.

Load-bearing premise

The load-bearing premise is that the hand-picked task-similarity measure, distance between the means of the best few sampled points in each task, ranks source tasks the way a genuinely useful transfer would; the adaptive weights, node potentials, and tree reconstructions all inherit whatever errors that ranking makes.

Editorial extensions

If this is right

  • A user can take any Bayesian optimizer and run it inside the promised leaf subspace, so improving the partition immediately improves any downstream acquisition function.
  • Because weights are recomputed online, the method can identify which source tasks resemble the target and progressively ignore misleading ones, which is the key to working under mixed transfer.
  • The tree's reconstruction gives a built-in warm-start for the first evaluation and a continuing correction mechanism, so the method does not need to know task similarity in advance.
  • The paper's runtime analysis indicates that back-propagation and reconstruction add only a small overhead relative to evaluation on expensive real-world problems, making the transfer mechanism affordable where evaluations dominate.
  • The paper's MCTS-transfer-PFN combination shows that the search-space transfer is orthogonal to the surrogate model, so its gains stack with stronger BO models.

Reading between the lines

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

  • Editorial inference: because the decay factor $\gamma$ controls how quickly source influence fades, MCTS-transfer is effectively a continuous interpolation between full transfer and from-scratch optimization; setting $\gamma$ per task, rather than fixing it at 0.99, could extend the method to settings where the target drifts over time.
  • Editorial inference: the similarity measure is chosen by sensitivity analysis on the same benchmarks used for evaluation, so an independent check would be to fix one distance and one weight rule on a held-out family of tasks and see whether the adaptive weights are still better than fixed equal weights.
  • Editorial inference: the tree represents the search space generically rather than through a geometric box, so the same recipe could be applied to categorical or conditional search spaces; a concrete test would be to run it on the hierarchical spaces the paper discusses as future work.
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 / 5 minor

Summary. The paper proposes MCTS-transfer, a search-space transfer method for Bayesian optimization. In a pre-learning stage it recursively partitions the search space by clustering source-task data and training binary classifiers, building an MCTS tree whose nodes correspond to subspaces. During optimization, node potentials are updated as a discounted, similarity-weighted combination of source and target sample means, UCB selects a leaf subspace for BO, and a Treeify procedure rebuilds subtrees that violate the left-better-than-right invariant. The authors evaluate MCTS-transfer-GP against non-transfer baselines (GP, LA-MCTS), three search-space transfer baselines (Box-GP, Ellipsoid-GP, Supervised-GP), and the surrogate-transfer method PFN, on BBOB, three real-world problems, Design-Bench, and HPOB, under both similar and mixed transfer settings, and report improved mean ranks. The paper includes a limitations section, hyperparameter sensitivity analyses in appendices, and a public code link.

Significance. If the empirical claims hold, MCTS-transfer would be a useful contribution to transfer Bayesian optimization: it is more flexible than existing search-space transfer methods, can be combined with any BO optimizer, and is evaluated on a wide range of benchmarks with code released. The adaptive weighting idea is interesting and the runtime analysis is helpful. However, the main claims are empirical, and the current evidence does not yet establish that the adaptive similarity weighting generalizes: the similarity measure and several hyperparameters are selected on the same benchmarks used for the headline results, and the weight curves used to illustrate 'identification' of similar tasks are produced by the very distance rule being validated. The UCB formula in Eq. (1)-(3) is also internally inconsistent. These issues are fixable, but they need to be addressed before the central claims can be accepted.

major comments (4)
  1. [Section 3.1 / Eq. (1)-(3)] Section 3.1 states that v_m in Eq. (1) is replaced by the potential value p_m from Eq. (2). Since p_m is already an average of objective values, and Eq. (3) later defines p_m as a weighted average of per-task means, the exploitation term in UCB becomes p_m/n_m, i.e., an average divided by the visit count, rather than the standard total-reward/n_m. This changes the exploration-exploitation balance and makes the node-selection rule ambiguous. Please clarify whether p_m is meant to be a total or an average and adjust Eq. (1) accordingly (e.g., use p_m directly, or define the total value as p_m·n_m).
  2. [Appendix D / Section 4.3 / Appendix B.1] The sensitivity analysis in Appendix D is conducted on LunarLander, RobotPush, and Rover under the mixed setting, and the same three problems appear in the main evaluation (Section 4.3, Figure 3b, and Appendix E, Figure 16). Appendix D further states that 'In mixed transfer real-world problems, KL divergence is more appropriate to the problem,' yet Appendix B.1 reports that the main experiments adopt the best-5-solutions distance. This means the similarity measure and several hyperparameters (γ, α, Cp, θ, classifier) are effectively selected on the test benchmarks, undermining the claim that MCTS-transfer can 'adaptively identify' similar source tasks in a general setting. Please provide a held-out validation of the similarity-measure choice or clearly restrict the claims to the tuned configuration.
  3. [Section 4.1 / Appendix G / Eq. (4)] The evidence for the claim that MCTS-transfer 'can extract the most similar source tasks and give them higher weights' consists of weight-change curves (Section 4.1, Figure 2(b); Appendix G). These weights are computed by Eq. (4) from exactly the best-N mean-distance rule whose validity is at issue, so the curves largely confirm the definition rather than demonstrating that the weighting improves transfer. An independent validation would compare optimization performance with adaptive weights against fixed/equal weights, or measure the correlation between assigned weights and task transfer usefulness.
  4. [Section 3.4 / Algorithm 2] Section 3.4 states that 'for node expansion, only samples of the target task are considered,' and Algorithm 2 reconstructs subtrees with the same expansion process. If Treeify also uses only target samples, then any subtree rebuilt during optimization discards the source-task data that shaped the original partition, which appears to contradict the paper's claim that the tree can 'retain historical information.' Please state explicitly whether source samples are included in the clustering/classification during Treeify, and if so, how they are combined with target samples.
minor comments (5)
  1. [Throughout] There are numerous typos: 'filed' should be 'field' (Section 2.4), 'ajusts' should be 'adjusts' (Section 4.1), 'pre-learnned' should be 'pre-learned' (Section 3.4), 'splitale' should be 'splittable' (Algorithm 1), 'tranfer' should be 'transfer' in several figure legends, 'Specailly' should be 'Especially' (Appendix B.3), 'he result' should be 'the result' (Appendix D), 'influencial' should be 'influential' (Appendix D), 'stategy' should be 'strategy' (Appendix G), 'distribition' should be 'distribution' (Appendix C.1), and 'Simialr' should be 'Similar' in Appendix E titles.
  2. [Appendix C.1] The sentence 'The first three methods are point-based measures and the last three are distribution-based methods' is inaccurate, as only five methods are listed and only the last two (Kendall coefficient and KL divergence) are distribution-based; please correct the count.
  3. [Section 3.2] The parameter N in the definition of Distance(x_i^*, x_T^*) is not specified in the main text; it first appears as '5 best solutions distance' in Appendix B.1. Please state the default value in the main text.
  4. [Section 3.2, Eq. (3)] The per-task means y_i,m and y_T,m are undefined when a node contains no samples from a source task or from the target, which can occur after subtree reconstruction and for nodes not on the selected path. Please specify the default handling (e.g., skip, zero, or smoothing).
  5. [Section 4.1] The text says 'Higher weights are assigned to datasets D(5,5) and D(-5,5)', but the three source datasets listed at the start of the section are D(5,5), D(5,-5), and D(-5,-5); D(-5,5) is not among them. Please correct the dataset labels.

Circularity Check

1 steps flagged · score 2.0 of 10

No load-bearing circularity; the only mild issue is a self-confirmatory sanity check of the adaptive-weight mechanism.

  1. self definitional [Section 3.2 (Eq. (4)) and Appendix G (weight change curves)]
    "To calculate wi, we measure the distance Distance(Di, DT ) between the i-th source task and the target task by Distance(x∗ i , x∗ T ), where x∗ i and x∗ T denote the mean of the best N sampled points of these two tasks, respectively. ... the weights of real-world problem and similar sphere problem exceed those of dissimilar sphere problem in most cases ... The results prove that the weight change strategy can prioritize similar source task data."

    The 'similar sphere' source tasks are generated by placing their optimum at the target's best-found solution, while the 'dissimilar sphere' uses the reflected optimum 1 - x*. The weight wi is itself defined by the distance between the mean of the best N target samples and the mean of the best N source samples. Under this definition, the similar sphere will almost by construction receive a higher weight than the reflected sphere, because its best-sample mean is geometrically closer to the target's best-sample mean. Thus the Appendix G curves confirm the definition of the similarity measure rather than independently validating that the measure identifies transfer-relevant tasks.

full rationale

The central contribution of MCTS-transfer is an empirical algorithm: MCTS with source-task-weighted node potentials, adaptive weights, and tree reconstruction. There is no theoretical derivation in which an output is equal to an input by construction. The main experimental claims are evaluated against external benchmarks (BBOB, Design-Bench, HPOB) and baseline algorithms, with code released, so the empirical comparison is self-contained and falsifiable. The one mild circular element is the paper's use of weight-change curves to 'prove' that the method identifies similar source tasks: the notion of similar versus dissimilar source tasks is constructed from the same distance measure that defines the weights, so the curves mostly restate the definition. This does not undermine the comparative benchmark results, which do not depend on that sanity check. A separate concern, not circularity, is that the sensitivity analysis in Appendix D selects hyperparameters and the similarity measure on the same real-world benchmarks used in the main evaluation; this is a test-set-selection risk rather than a logical reduction of the method's claims to its inputs.

Assumptions & free parameters 7 free parameters · 4 assumptions · 0 invented entities

The method rests on several heuristic modeling choices: a similarity proxy, a tree-partitioning scheme, and a set of hyperparameters tuned on the evaluation benchmarks. No new physical or ontological entities are introduced. The central mechanism depends more on empirical tuning than on a theory with stated assumptions.

free parameters (7)
  • gamma (decay factor) = 0.99
    Controls how quickly source-task influence decays via gamma^(t-1) in Eq. (3). Chosen by sensitivity analysis on LunarLander, RobotPush, and Rover in Appendix D, and used in the main experiments.
  • alpha (important source task ratio) = 0.5
    Determines the fraction of source tasks receiving high weight in Eq. (4). Selected via sensitivity analysis (Appendix D), with values 0.1, 0.5, 1.0 tested.
  • Cp (UCB exploration factor) = 0.1
    Exploration constant in Eq. (1), inherited from LA-MCTS. Sensitivity analysis in Appendix D shows performance varies with Cp.
  • theta (splitting threshold) = 10
    Minimum number of samples required before a node can be split. Inherited from LA-MCTS; sensitivity analysis shows theta=3, 10, 100 affect tree depth and performance.
  • N (best-N for similarity) = 5
    Number of best target/source samples whose mean is used as the similarity distance in Section 3.2. Chosen as default among several measures compared in Appendix C/D.
  • Weight assignment strategy = linear change (Eq. 7)
    Selected among linear, exponential, and all-one strategies based on sensitivity analysis on the three real-world problems.
  • Binary classifier = SVM (rbf) for Sphere2D/BBOB/HPOB; Logistic Regression for real-world
    The classifier defines good/bad boundaries during space partitioning. Sensitivity analysis in Appendix D shows classification choice changes results; the choice is made per problem.
assumptions (4)
  • domain assumption Source tasks and the target task share common characteristics that make a shared partition of the search space informative.
    Core transfer-learning premise stated in Section 2.2 and used throughout the pre-learning and optimization stages. If false, the pre-learned tree provides no useful warm-start.
  • ad hoc to paper The mean of the best N sampled points is a valid proxy for task similarity.
    Eq. (4) and Section 3.2 use Distance(x_i^*, x_T^*) where x^* is the mean of best N points. Appendix C shows different measures give different results, so this is a heuristic choice rather than a justified metric.
  • domain assumption A binary partition by k-means clusters plus a classifier separates good from bad regions well enough for UCB-guided selection.
    Section 3.1 and Algorithm 1 rely on this for node expansion and reconstruction. Sensitivity analysis shows performance depends on classifier choice.
  • ad hoc to paper The tree invariant (left child potential >= right child potential) can be restored by deleting and rebuilding subtrees without losing important information.
    Treeify in Section 3.3 assumes reconstruction improves the partition. No analysis is given for information loss or convergence effects.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Monte Carlo Tree Search based Space Transfer for Black-box Optimization." pith.science (2026). https://pith.science/paper/KHC5H7WD

@misc{pith2026241207186,
  author       = {Pith},
  title        = {Pith review of: Monte Carlo Tree Search based Space Transfer for Black-box Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KHC5H7WD}},
  note         = {Machine review of arXiv:2412.07186}
}
read the original abstract

Bayesian optimization (BO) is a popular method for computationally expensive black-box optimization. However, traditional BO methods need to solve new problems from scratch, leading to slow convergence. Recent studies try to extend BO to a transfer learning setup to speed up the optimization, where search space transfer is one of the most promising approaches and has shown impressive performance on many tasks. However, existing search space transfer methods either lack an adaptive mechanism or are not flexible enough, making it difficult to efficiently identify promising search space during the optimization process. In this paper, we propose a search space transfer learning method based on Monte Carlo tree search (MCTS), called MCTS-transfer, to iteratively divide, select, and optimize in a learned subspace. MCTS-transfer can not only provide a well-performing search space for warm-start but also adaptively identify and leverage the information of similar source tasks to reconstruct the search space during the optimization process. Experiments on synthetic functions, real-world problems, Design-Bench and hyper-parameter optimization show that MCTS-transfer can demonstrate superior performance compared to other search space transfer methods under different settings. Our code is available at \url{https://github.com/lamda-bbo/mcts-transfer}.

Figures

Figures reproduced from arXiv: 2412.07186 by the authors.

Figure 1
Figure 1. The workflow of MCTS-transfer. In pre-learning stage, MCTS-transfer builds the tree [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 3
Figure 3. Comparison between MCTS-transfer and other algorithms on BBOB and real-world [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Comparison between MCTS-transfer and other algorithms on Design-Bench and HPOB. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figures from the paper (17 more)
Figure 5
Figure 5. Figure 5: Time cost proportion of evaluation, backpropogation and reconstruction of MCTS-transfer. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Experimental results on mixed real world [PITH_FULL_IMAGE:figures/full_fig_p018_6.png]
Figure 7
Figure 7. Figure 7: Sensitivity Analysis of Local and Global Modeling Approaches [PITH_FULL_IMAGE:figures/full_fig_p019_7.png]
Figure 8
Figure 8. Figure 8: Sensitivity Analysis of Similarity Measures [PITH_FULL_IMAGE:figures/full_fig_p019_8.png]
Figure 9
Figure 9. Figure 9: Sensitivity Analysis of Weight Change Strategy [PITH_FULL_IMAGE:figures/full_fig_p020_9.png]
Figure 10
Figure 10. Figure 10: Sensitivity Analysis of Decay Factor γ Important source task ratio α In linear-change strategy, we can set α to determine the ratio of important source tasks that have high weights. We choose α = 10%, 50% and 100% and he result is shown in [PITH_FULL_IMAGE:figures/fu…
Figure 11
Figure 11. Figure 11: Sensitivity Analysis of Important Source Task Ratio [PITH_FULL_IMAGE:figures/full_fig_p021_11.png]
Figure 12
Figure 12. Figure 12: Sensitivity Analysis of Exploration Factor [PITH_FULL_IMAGE:figures/full_fig_p021_12.png]
Figure 13
Figure 13. Figure 13: Sensitivity Analysis of The Splitting Threshold [PITH_FULL_IMAGE:figures/full_fig_p021_13.png]
Figure 14
Figure 14. Figure 14: Sensitivity Analysis of Binary Classifier [PITH_FULL_IMAGE:figures/full_fig_p022_14.png]
Figure 15
Figure 15. Figure 15: Evaluations of MCTS-transfer and other algorithms on BBOB [PITH_FULL_IMAGE:figures/full_fig_p023_15.png]
Figure 16
Figure 16. Figure 16: Evaluations of MCTS-transfer and other algorithms on Real-world Problems [PITH_FULL_IMAGE:figures/full_fig_p024_16.png]
Figure 17
Figure 17. Figure 17: Evaluations of MCTS-transfer and other algorithms on Design-Bench [PITH_FULL_IMAGE:figures/full_fig_p024_17.png]
Figure 18
Figure 18. Figure 18: Evaluations of MCTS-transfer and other algorithms on HPOB in similar setting 25 [PITH_FULL_IMAGE:figures/full_fig_p025_18.png]
Figure 19
Figure 19. Figure 19: Evaluations of MCTS-transfer and other algorithms on HPOB in mixed setting 26 [PITH_FULL_IMAGE:figures/full_fig_p026_19.png]
Figure 20
Figure 20. Figure 20: Time cost of different components in MCTS-transfer. The top row: the time proportion [PITH_FULL_IMAGE:figures/full_fig_p027_20.png]
Figure 21
Figure 21. Figure 21: shows that the weights of real-world problem and similar sphere problem exceed those of dissimilar sphere problem in most cases, regardless of any inconsistencies in initialization. The results prove that the weight change strategy can prioritize similar source task d…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Policy Guided Tree Search for Enhanced LLM Reasoning

    cs.LG 2025-02 conditional novelty 5.0 of 10

    PGTS trains a graph-transformer policy via PPO to guide tree search over LLM reasoning steps, reporting improved accuracy and lower token use than chain-of-thought and MCTS baselines.

Reference graph

Works this paper leans on

71 extracted references · 59 canonical work pages · cited by 1 Pith paper

  1. [1]

    Finite-time analysis of the multiarmed bandit problem

    Peter Auer, Nicolò Cesa-Bianchi, and Paul Fischer. Finite-time analysis of the multiarmed bandit problem. Machine learning, 47(2):235–256, 2002

    work page 2002

  2. [2]

    Transfer learning for Bayesian optimization: A survey

    Tianyi Bai, Yang Li, Yu Shen, Xinyi Zhang, Wentao Zhang, and Bin Cui. Transfer learning for Bayesian optimization: A survey. arXiv:2302.05927, 2023

    arXiv 2023

  3. [3]

    Random search for hyper-parameter optimization

    James Bergstra and Yoshua Bengio. Random search for hyper-parameter optimization. Journal of Machine Learning Research, 13:281–305, 2012

    work page 2012

  4. [4]

    Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges

    Bernd Bischl, Martin Binder, Michel Lang, Tobias Pielok, Jakob Richter, Stefan Coors, Janek Thomas, Theresa Ullmann, Marc Becker, Anne-Laure Boulesteix, Difan Deng, and Marius Lindauer. Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges. WIREs Data. Mining. Knowl. Discov., 13(2), 2023

    work page 2023

  5. [5]

    Lucas, Peter I

    Cameron Browne, Edward Jack Powley, Daniel Whitehouse, Simon M. Lucas, Peter I. Cowling, Philipp Rohlfshagen, Stephen Tavener, Diego Perez Liebana, Spyridon Samothrakis, and Simon Colton. A survey of Monte Carlo tree search methods. IEEE Transactions on Computational Intelligence and AI in Games, 4(1):1–43, 2012

    work page 2012

  6. [6]

    Transfer Learning for Bayesian Optimization on Heterogeneous Search Spaces

    Zhou Fan, Xinran Han, and Zi Wang. Transfer learning for bayesian optimization on heteroge- neous search spaces. arXiv preprint arXiv:2309.16597, 2023

    work page Pith review arXiv 2023

  7. [7]

    Initializing Bayesian hyperparam- eter optimization via meta-learning

    Matthias Feurer, Jost Tobias Springenberg, and Frank Hutter. Initializing Bayesian hyperparam- eter optimization via meta-learning. In Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI’15), pages 1128–1135, Austin, TX, 2015

    work page 2015

  8. [8]

    Peter I. Frazier. A tutorial on Bayesian optimization. arXiv:1807.02811, 2018

    arXiv 2018

Show all 71 references
  1. [9]

    Real-parameter black-box optimization benchmarking 2009: Noiseless functions definitions

    Nikolaus Hansen, Steffen Finck, Raymond Ros, and Anne Auger. Real-parameter black-box optimization benchmarking 2009: Noiseless functions definitions. Technical Report RR-6829, INRIA, 2009

    2009

  2. [10]

    End-to-end learning of deep kernel acquisition functions for Bayesian opti- mization

    Tomoharu Iwata. End-to-end learning of deep kernel acquisition functions for Bayesian opti- mization. arXiv:2111.00639, 2021

    2021 arXiv

  3. [11]

    Transfer learning for Bayesian HPO with end-to-end meta-features, 2022

    Hadi Samer Jomaa, Sebastian Pineda Arango, Lars Schmidt-Thieme, and Josif Grabocka. Transfer learning for Bayesian HPO with end-to-end meta-features, 2022

    2022

  4. [12]

    Jones, Matthias Schonlau, and William J

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

    1998

  5. [13]

    A flexible transfer learning framework for Bayesian optimization with convergence guarantee

    Tinu Theckel Joy, Santu Rana, Sunil Gupta, and Svetha Venkatesh. A flexible transfer learning framework for Bayesian optimization with convergence guarantee. Expert Syst. Appl., 115: 656–672, 2019

    2019

  6. [14]

    Monte carlo tree search in continuous spaces using voronoi optimistic optimization with regret bounds

    Beomjoon Kim, Kyungjae Lee, Sungbin Lim, Leslie Pack Kaelbling, and Tomás Lozano-Pérez. Monte carlo tree search in continuous spaces using voronoi optimistic optimization with regret bounds. In The 34th AAAI Conference on Artificial Intelligence (AAAI’20), pages 9916–9924, New...

    2020

  7. [15]

    Harold J. Kushner. A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise. Journal of Basic Engineering, 86(1):97–106, 1964

    1964

  8. [16]

    Hy- perparameter learning via distributional transfer

    Ho Chung Leon Law, Peilin Zhao, Leung Sing Chan, Junzhou Huang, and Dino Sejdinovic. Hy- perparameter learning via distributional transfer. In Advances in Neural Information Processing Systems 32 (NeurIPS’18), pages 6801–6812, Vancouver, Canada, 2018

    2018

  9. [17]

    Transfer learning based search space design for hyperparameter tuning

    Yang Li, Yu Shen, Huaijun Jiang, Tianyi Bai, Wentao Zhang, Ce Zhang, and Bin Cui. Transfer learning based search space design for hyperparameter tuning. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD’22), pages 967–977, Washington,...

    2022

  10. [18]

    Blaschko

    Xingchen Ma and Matthew B. Blaschko. Additive tree-structured covariance function for conditional parameter spaces in bayesian optimization. In Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS’20, pages 1015–1025, Palermo, Italy, 2020

    2020

  11. [19]

    Azalia Mirhoseini, Anna Goldie, Mustafa Yazgan, Joe W. J. Jiang, Ebrahim M. Songhori, Shen Wang, Young-Joon Lee, Eric Johnson, Omkar Pathak, Sungmin Bae, Azade Nazi, Jiwoo Pak, 11 Andy Tong, Kavya Srinivasa, William Hang, Emre Tuncer, Anand Babu, Quoc V . Le, James Laudon, Ric...

    2004 arXiv

  12. [20]

    Moss, David S

    Henry B. Moss, David S. Leslie, and Paul Rayson. Mumbo: Multi-task max-value Bayesian optimization. In Proceedings of the 31th Machine Learning and Knowledge Discovery in Databases: European Conference (ECML/PKDD’20), pages 447–462, Ghent, Belgium, 2020

    2020

  13. [21]

    Pfns4bo: In-context learning for Bayesian optimization

    Samuel Müller, Matthias Feurer, Noah Hollmann, and Frank Hutter. Pfns4bo: In-context learning for Bayesian optimization. In Proceedings of the 40th International Conference on Machine Learning (ICML’23), pages 25444–25470, Honolulu, HI, 2023

    2023

  14. [22]

    Optimistic optimization of a deterministic function without the knowledge of its smoothness

    Rémi Munos. Optimistic optimization of a deterministic function without the knowledge of its smoothness. In Advances in Neural Information Processing Systems 24 (NeurIPS’11), pages 783–791, Granada, Spain, 2011

    2011

  15. [23]

    Learning search spaces for Bayesian optimization: Another view of hyperparameter transfer learning

    Valerio Perrone and Huibin Shen. Learning search spaces for Bayesian optimization: Another view of hyperparameter transfer learning. In Advances in Neural Information Processing Systems 32 (NeurIPS’19), pages 12751–12761, Vancouver, Canada, 2019

    2019

  16. [24]

    Jomaa, Martin Wistuba, and Josif Grabocka

    Sebastian Pineda-Arango, Hadi S. Jomaa, Martin Wistuba, and Josif Grabocka. HPO-B: A large-scale reproducible benchmark for black-box HPO based on openml. In Advances in Neural Information Processing Systems 34 (NeurIPS’21), Virtual, 2021

    2021

  17. [25]

    Selecting optimal source for transfer learning in Bayesian optimisation

    Anil Ramachandran, Sunil Gupta, Santu Rana, and Svetha Venkatesh. Selecting optimal source for transfer learning in Bayesian optimisation. In Proceedings of the 15th Pacific Rim International Conference on Artificial Intelligence (PRICAI’18), pages 42–56, Nanjing, China, 2018

    2018

  18. [26]

    Carl Edward Rasmussen and Christopher K. I. Williams. Gaussian Processes for Machine Learning. The MIT Press, 2006

    2006

  19. [27]

    Esteban Real, Alok Aggarwal, Yanping Huang, and Quoc V . Le. Regularized evolution for image classifier architecture search. In Proceedings of the 33rd AAAI Conference on Artificial Intelligence (AAAI’19), pages 4780–4789, Honolulu, HI, 2019

    2019

  20. [28]

    Scalable hyperparameter op- timization with products of Gaussian process experts

    Nicolas Schilling, Martin Wistuba, and Lars Schmidt-Thieme. Scalable hyperparameter op- timization with products of Gaussian process experts. In Proceedings of the 27th Machine Learning and Knowledge Discovery in Databases: European Conference (ECML/PKDD’16), pages 199–214, Ri...

    2016

  21. [29]

    Adams, and Nando de Freitas

    Bobak Shahriari, Kevin Swersky, Ziyu Wang, Ryan P. Adams, and Nando de Freitas. Taking the human out of the loop: A review of Bayesian optimization. Proceedings of the IEEE, 104 (1):148–175, 2016

    2016

  22. [30]

    Macro placement by wire-mask-guided black-box optimization

    Yunqi Shi, Ke Xue, Lei Song, and Chao Qian. Macro placement by wire-mask-guided black-box optimization. In Advances in Neural Information Processing Systems 36 (NeurIPS’23), New Orleans, LA, 2023

    2023

  23. [31]

    Regret bounds for transfer learning in Bayesian optimisation

    Alistair Shilton, Sunil Gupta, Santu Rana, and Svetha Venkatesh. Regret bounds for transfer learning in Bayesian optimisation. In Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS’17), pages 307–315, Fort Lauderdale, FL, 2017

    2017

  24. [32]

    David Silver, Aja Huang, Chris J. Maddison, Arthur Guez, Laurent Sifre, George van den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Vedavyas Panneershelvam, Marc Lanctot, Sander Dieleman, Dominik Grewe, John Nham, Nal Kalchbrenner, Ilya Sutskever, Timothy P. Lillicrap,...

    2016

  25. [33]

    Lillicrap, Karen Simonyan, and Demis Hassabis

    David Silver, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, Thore Graepel, Timothy P. Lillicrap, Karen Simonyan, and Demis Hassabis. Mastering chess and shogi by self-play with a general reinfo...

    2017 arXiv

  26. [34]

    Lillicrap, Fan Hui, Laurent Sifre, George van den Driessche, Thore Graepel, and Demis Hassabis

    David Silver, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai, Adrian Bolton, Yutian Chen, Timothy P. Lillicrap, Fan Hui, Laurent Sifre, George van den Driessche, Thore Graepel, and Demis Hassabis. Maste...

    2017

  27. [35]

    Monte Carlo tree search based variable selection for high dimensional Bayesian optimization

    Lei Song, Ke Xue, Xiaobin Huang, and Chao Qian. Monte Carlo tree search based variable selection for high dimensional Bayesian optimization. In Advances in Neural Information Processing Systems 35 (NeurIPS’22), New Orleans, LA, 2022

    2022

  28. [36]

    Kakade, and Matthias W

    Niranjan Srinivas, Andreas Krause, Sham M. Kakade, and Matthias W. Seeger. Information- theoretic regret bounds for Gaussian process optimization in the bandit setting. IEEE Transac- tions on Information Theory, 58(5):3250–3265, 2012

    2012

  29. [37]

    Multi-task Bayesian optimization

    Kevin Swersky, Jasper Snoek, and Ryan Prescott Adams. Multi-task Bayesian optimization. In Advances in Neural Information Processing Systems 26 (NeurIPS’13), pages 2004–2012, Lake Tahoe, NV , 2013

    2004

  30. [38]

    Transfer learning with Gaussian processes for Bayesian optimization

    Petru Tighineanu, Kathrin Skubch, Paul Baireuther, Attila Reiss, Felix Berkenkamp, and Julia Vinogradska. Transfer learning with Gaussian processes for Bayesian optimization. In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS...

    2022

  31. [39]

    Design-bench: Bench- marks for data-driven offline model-based optimization

    Brandon Trabucco, Xinyang Geng, Aviral Kumar, and Sergey Levine. Design-bench: Bench- marks for data-driven offline model-based optimization. InProceedings of the 39th International Conference on Machine Learning (ICML’22), pages 21658–21676, Baltimore, MD, 2022

    2022

  32. [40]

    Fröhlich, Kirsten Fischer, Andreas Doerr, Stefan Falkner, Frank Hutter, and Christian Daniel

    Michael V olpp, Lukas P. Fröhlich, Kirsten Fischer, Andreas Doerr, Stefan Falkner, Frank Hutter, and Christian Daniel. Meta-learning acquisition functions for transfer learning in Bayesian optimization. In Proceedings of the 8th International Conference on Learning Representat...

    2019

  33. [41]

    Sample-efficient neural architecture search by learning action space

    Linnan Wang, Saining Xie, Teng Li, Rodrigo Fonseca, and Yuandong Tian. Sample-efficient neural architecture search by learning action space. arXiv:1906.06832, 2019

    1906 arXiv

  34. [42]

    Alphax: exploring neural architectures with deep neural networks and monte carlo tree search

    Linnan Wang, Yiyang Zhao, Yuu Jinnai, Yuandong Tian, and Rodrigo Fonseca. Alphax: exploring neural architectures with deep neural networks and monte carlo tree search. arXiv:1903.11059, 2019

    1903 arXiv

  35. [43]

    Learning search space partition for black-box optimization using monte carlo tree search

    Linnan Wang, Rodrigo Fonseca, and Yuandong Tian. Learning search space partition for black-box optimization using monte carlo tree search. In Advances in Neural Information Processing Systems 33 (NeurIPS’20), pages 19511–19522, Virtual, 2020

    2020

  36. [44]

    Batched large-scale Bayesian optimization in high-dimensional spaces

    Zi Wang, Clement Gehring, Pushmeet Kohli, and Stefanie Jegelka. Batched large-scale Bayesian optimization in high-dimensional spaces. In Proceedings of 21st International Conference on Artificial Intelligence and Statistics (AISTATS’18), pages 745–754, Playa Blanca, Spain, 2018

    2018

  37. [45]

    Meta-learning hyperparameter performance prediction with neural processes

    Ying Wei, Peilin Zhao, and Junzhou Huang. Meta-learning hyperparameter performance prediction with neural processes. In Proceedings of the 38th International Conference on Machine Learning (ICML’21), pages 11058–11067, Virtual, 2021

    2021

  38. [46]

    Wilson, Riccardo Moriconi, Frank Hutter, and Marc Peter Deisenroth

    James T. Wilson, Riccardo Moriconi, Frank Hutter, and Marc Peter Deisenroth. The reparame- terization trick for acquisition functions. arXiv:1712.00424, 2017

    2017 arXiv

  39. [47]

    Few-shot Bayesian optimization with deep kernel sur- rogates

    Martin Wistuba and Josif Grabocka. Few-shot Bayesian optimization with deep kernel sur- rogates. In Proceedings of the 9th International Conference on Learning Representations (ICLR’21), Virtual, 2021

    2021

  40. [48]

    Learning hyperparameter optimization initializations

    Martin Wistuba, Nicolas Schilling, and Lars Schmidt-Thieme. Learning hyperparameter optimization initializations. In Proceedings of the 2nd IEEE International Conference on Data Science and Advanced Analytics (DSAA’15), pages 1–10, Paris, France, 2015

    2015

  41. [49]

    Hyperparameter search space pruning - A new component for sequential model-based hyperparameter optimization

    Martin Wistuba, Nicolas Schilling, and Lars Schmidt-Thieme. Hyperparameter search space pruning - A new component for sequential model-based hyperparameter optimization. In Proceedings of the 26th Machine Learning and Knowledge Discovery in Databases: European Conference (ECML...

    2015

  42. [50]

    Two-stage transfer surrogate model for automatic hyperparameter optimization

    Martin Wistuba, Nicolas Schilling, and Lars Schmidt-Thieme. Two-stage transfer surrogate model for automatic hyperparameter optimization. InProceedings of the 27th Machine Learning and Knowledge Discovery in Databases: European Conference (ECML/PKDD’16), pages 199– 214, Riva d...

    2016

  43. [51]

    Scalable Gaussian process-based transfer surrogates for hyperparameter optimization

    Martin Wistuba, Nicolas Schilling, and Lars Schmidt-Thieme. Scalable Gaussian process-based transfer surrogates for hyperparameter optimization. Machine Learning, 107:43–78, 2017

    2017

  44. [52]

    Practical two-step lookahead bayesian optimization

    Jian Wu and Peter Frazier. Practical two-step lookahead bayesian optimization. In Advances in Neural Information Processing Systems 32 (NeurIPS’19), 2019. 13

    2019

  45. [53]

    Carlberg, Neil Walton, and Kody J

    Shangda Yang, Vitaly Zankin, Maximilian Balandat, Stefan Scherer, Kevin T. Carlberg, Neil Walton, and Kody J. H. Law. Accelerating look-ahead in bayesian optimization: Multilevel monte carlo is all you need. In Proceedings of the 41st International Conference on Machine Learni...

    2024

  46. [54]

    Eagle strategy using lévy walk and firefly algorithms for stochastic optimization

    Xin-She Yang and Suash Deb. Eagle strategy using lévy walk and firefly algorithms for stochastic optimization. In Proceedings of the 4th Nature Inspired Cooperative Strategies for Optimization (NICSO’10), pages 101–111, Granada, Spain, 2010

    2010

  47. [55]

    Taking human out of learning applications: A survey on automated machine learning

    Quanming Yao, Mengshuo Wang, Hugo Jair Escalante, Isabelle Guyon, Yi-Qi Hu, Yu-Feng Li, Wei-Wei Tu, Qiang Yang, and Yang Yu. Taking human out of learning applications: A survey on automated machine learning. arXiv:1810.13306, 2018

    arXiv

  48. [56]

    Barret Zoph and Quoc V . Le. Neural architecture search with reinforcement learning. In Proceedings of the 5th International Conference on Learning Representations (ICLR’17) , Toulon, France, 2017. 14 A Treeify Pseudocode To maintain the property that the potential of the left...

    2017

  49. [57]

    Guidelines: • The answer NA means that the abstract and introduction do not include the claims made in the paper

    Claims Question: Do the main claims made in the abstract and introduction accurately reflect the paper’s contributions and scope? Answer: [Yes] Justification: Both the abstract and introduction accurately reflect the paper’s contributions and scope. Guidelines: • The answer NA...

  50. [58]

    Limitations

    Limitations Question: Does the paper discuss the limitations of the work performed by the authors? Answer: [Yes] Justification: The limitations are proposed in the Conclusion section. Guidelines: • The answer NA means that the paper has no limitation while the answer No means ...

  51. [59]

    Guidelines: • The answer NA means that the paper does not include theoretical results

    Theory Assumptions and Proofs Question: For each theoretical result, does the paper provide the full set of assumptions and a complete (and correct) proof? Answer: [NA] 29 Justification: The paper does not include theoretical results. Guidelines: • The answer NA means that the...

  52. [60]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experimental Result Reproducibility Question: Does the paper fully disclose all the information needed to reproduce the main ex- perimental results of the paper to the extent that it affects the main claims and/or conclusions of the paper (regardless of whether the code and da...

  53. [61]

    Guidelines: • The answer NA means that paper does not include experiments requiring code

    Open access to data and code Question: Does the paper provide open access to the data and code, with sufficient instruc- tions to faithfully reproduce the main experimental results, as described in supplemental material? 30 Answer: [Yes] Justification: We will submit code, dat...

  54. [62]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experimental Setting/Details Question: Does the paper specify all the training and test details (e.g., data splits, hyper- parameters, how they were chosen, type of optimizer, etc.) necessary to understand the results? Answer: [Yes] Justification: The detailed information is p...

  55. [63]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experiment Statistical Significance Question: Does the paper report error bars suitably and correctly defined or other appropriate information about the statistical significance of the experiments? Answer: [Yes] Justification: The error bars are already shown in figures in pap...

  56. [64]

    Guidelines: • The answer NA means that the paper does not include experiments

    Experiments Compute Resources Question: For each experiment, does the paper provide sufficient information on the com- puter resources (type of compute workers, memory, time of execution) needed to reproduce the experiments? Answer: [No] Justification: We didn’t provide comput...

  57. [65]

    Guidelines: • The answer NA means that the authors have not reviewed the NeurIPS Code of Ethics

    Code Of Ethics Question: Does the research conducted in the paper conform, in every respect, with the NeurIPS Code of Ethics https://neurips.cc/public/EthicsGuidelines? Answer: [Yes] Justification: Our research conform with the NeurIPS Code of Ethics. Guidelines: • The answer ...

  58. [66]

    Guidelines: • The answer NA means that there is no societal impact of the work performed

    Broader Impacts Question: Does the paper discuss both potential positive societal impacts and negative societal impacts of the work performed? Answer: [Yes] Justification: We describe it in introduction part. Guidelines: • The answer NA means that there is no societal impact o...

  59. [67]

    Guidelines: • The answer NA means that the paper poses no such risks

    Safeguards Question: Does the paper describe safeguards that have been put in place for responsible release of data or models that have a high risk for misuse (e.g., pretrained language models, image generators, or scraped datasets)? Answer: [NA] Justification: The paper poses...

  60. [68]

    Guidelines: • The answer NA means that the paper does not use existing assets

    Licenses for existing assets Question: Are the creators or original owners of assets (e.g., code, data, models), used in the paper, properly credited and are the license and terms of use explicitly mentioned and properly respected? Answer: [Yes] Justification: We correctly cit...

  61. [69]

    Guidelines: • The answer NA means that the paper does not release new assets

    New Assets Question: Are new assets introduced in the paper well documented and is the documentation provided alongside the assets? Answer: [NA] Justification: The paper does not release new assets. Guidelines: • The answer NA means that the paper does not release new assets. ...

  62. [70]

    Guidelines: • The answer NA means that the paper does not involve crowdsourcing nor research with human subjects

    Crowdsourcing and Research with Human Subjects Question: For crowdsourcing experiments and research with human subjects, does the paper include the full text of instructions given to participants and screenshots, if applicable, as well as details about compensation (if any)? A...

  63. [71]

    Guidelines: • The answer NA means that the paper does not involve crowdsourcing nor research with human subjects

    Institutional Review Board (IRB) Approvals or Equivalent for Research with Human Subjects Question: Does the paper describe potential risks incurred by study participants, whether such risks were disclosed to the subjects, and whether Institutional Review Board (IRB) approvals...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.