Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Semivalue-based data valuation is arbitrary and gameable

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Pith's one-line read Semivalue-based data valuation is arbitrary and gameable.

desk verdict The arbitrariness argument is solid, but the gameability results are trivial or buggy as written. read the letter →

arxiv 2506.12619 v1 pith:BTZKXUJI submitted 2025-06-14 cs.LG cs.GT

classification cs.LGcs.GT MSC 91A12
keywords datavaluationsemivalueShapleyvalueBanzhafutilityunderspecificationarbitrarinessgameabilitycooperativegametheory
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

Semivalue methods such as Data Shapley and Data Banzhaf promise a principled, axiom-backed way to assign credit to individual training examples. This paper argues that the promise cannot survive contact with practice: the utility function that scores every subset of data is under-specified, and small but defensible changes—how tiny training coalitions are handled, whether the score is monotonically rescaled, or how false positives are weighted—can shift individual payouts by multiple average contributors and reorder ranks enough to change which points a low-value filter drops. Because many of these utilities are observationally equivalent at the level of model selection, a bad-faith valuator can choose the one that favors a preferred group at low extra cost, while a good-faith valuator has no principled basis to single one out. The paper formalizes this as ambiguity sets and favorability metrics, proves that several ambiguity classes are gameable, and demonstrates large value and rank shifts on real benchmark datasets. If the claim holds, semivalue-based valuations are contestable instruments whose outputs should never be reported without a justification of the utility choice that produced them.

What carries the argument

The central object is the semivalue, $\psi_j(U,w)=\sum_{S\subseteq D\setminus\{z_j\}} w_{|S|}[U(S\cup\{z_j\})-U(S)]$, a weighted average of a datapoint's marginal contributions across all coalitions; Data Shapley, Data Banzhaf, and leave-one-out are instances that differ only in the weights $w_{|S|}$. The argument's machinery is the ambiguity set: a family of utility functions that are observationally equivalent at the level of model selection but assign different scores to coalitions. Three such sets drive the analysis—small-data algorithmic ambiguity (behavior on tiny coalitions is undefined, so fallbacks are defensible), score-transformation ambiguity (monotone rescaling preserves model choice), and cost-ratio ambiguity (net benefit with unknown false-positive versus false-negative costs). Favorability metrics (aggregate payout, rank, filter survival) convert a value vector into a concrete outcome for a preferred group, and the gameability definitions count adversarial cost in extra utility evaluations beyond one semivalue computation. Algorithms that exploit semivalue linearity and the finite range of plausible behaviors locate the most favorable utility in these ambiguity sets with only polynomial extra evaluations.

What would settle it

Run a fixed benchmark where Data Shapley payouts are computed for two utilities that differ only in the small-coalition fallback threshold, say $0.05|D|$ and $0.1|D|$, and repeat across many datasets. If the median absolute payout change stays well below one average contributor and the bottom-10% filter memberships are identical, the arbitrariness claim for that ambiguity class would be contradicted.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that every semivalue-based data valuation silently commits to a counterfactual model that the learning task does not determine. For a fixed dataset and model, three families of equally defensible choices—the fallback behavior of the learning algorithm on coalitions below a size threshold, monotone transformations of the performance score, and the false-positive cost ratio in a net-benefit metric—produce substantially different semivalue vectors. These choices are invisible to model selection, so the valuation pipeline can be driven toward a preferred group without any visible distortion of the model itself. The paper further shows that low-cost algorithms can search these ambiguity sets for the utility that maximizes a chosen group's aggregate value, rank, or filter survival, and that empirical Shapley and Banzhaf payouts on standard datasets vary by several average units under such perturbations. The conclusion is that the fairness and objectivity attributed to semivalues are contingent on utility choices that are neither fixed by the data nor settled by any axiom.

Load-bearing premise

The load-bearing premise is that the utility functions inside each ambiguity set are all genuinely reasonable; if for a given task only one utility specification is actually correct, the valuations are no longer arbitrary and the gameability exploit disappears.

Editorial extensions

If this is right

  • Data Shapley and Data Banzhaf payouts should not be reported as point estimates; a single utility choice can move an individual's payout by several average contributors, so any report should include the sensitivity range over plausible utilities.
  • Decisions that use semivalues for contributor compensation, dataset acquisition, or low-value filtering inherit the arbitrariness; two equally defensible specifications can disagree about whose data is kept or credited.
  • A bad-faith valuator can systematically favor a target group by picking a defensible utility, so transparency about the selected utility is necessary but not sufficient for fairness claims.
  • Leave-one-out valuation is comparatively robust to small-coalition and score-transformation ambiguities because it depends only on full and leave-one-out subsets, but it remains sensitive to the choice of performance metric.
  • The same gameability algorithms can be run by a good-faith evaluator as a robustness audit: report the worst- and best-case valuations over an ambiguity set alongside the chosen specification.

Reading between the lines

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

  • The paper's gameability cost is measured as utility evaluations beyond one semivalue computation, which is already exponential in the dataset size; under an end-to-end cost measure, the 'low-cost' claim would need a separate argument that an adversary can beat the exponential semivalue computation itself.
  • The arbitrariness argument transfers naturally to feature-based Shapley values and other attribution schemes: any counterfactual that is not uniquely defined creates a similar ambiguity set, and the same favorability metrics could audit interpretability claims.
  • A practical guardrail suggested by the structure of Proposition 3 is that for linear utility families such as net benefit, the favorability extremes over a parameter interval occur at the interval's endpoints, so an auditor can bound adversarial advantage by evaluating only two utilities.
  • If regulators or platforms require pre-registration of the utility function, the exploit narrows, but the epistemic problem remains because the ambiguity set is in principle unbounded when all monotone score transformations are admitted.
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

3 major / 4 minor

Summary. The manuscript argues that semivalue-based data valuation (Shapley, Banzhaf, LOO) is fundamentally underspecified because the utility function V∘A involves modeling choices that the learning task does not determine. It formalizes three ambiguity families: small-data algorithmic behavior (Definitions 2–3), monotone score transformations (Definition 4), and cost-ratio choices (Definition 5). It introduces favorability metrics (aggregate value, rank, filter survival) and a range measure to quantify how much these choices can shift individual or group valuations. Empirically, it reports large payout and rank shifts on five datasets for both Shapley and Banzhaf values. The paper also defines exact and (ε,δ)-gameability (Definitions 6–7) and claims Propositions 1–3, with Algorithms 2–5, showing that a biased valuator can efficiently find utility specifications that favor a target group. It concludes with ethical implications and a burden-of-justification argument for semivalue-based valuation.

Significance. The arbitrariness analysis is a useful and timely contribution: formalizing concrete ambiguity sets and measuring their effect on payout, rank, and filter-survival metrics provides a clear framework for an often-invoked but underdeveloped critique of Shapley-style data valuation. The monotonic-transformation and small-coalition arguments are mathematically straightforward, and the empirical sensitivity results are plausible as illustrations. The gameability part is intended to sharpen the stakes, but in its current form it is not established: the exact-gameability definition trivializes constant-size candidate sets, and the small-cardinality gaming algorithm and its proof contain an incorrect coefficient. These are load-bearing issues for the 'gameable' claim in the abstract. The arbitrariness contribution is not invalidated by these problems, but the formal gameability claims need substantive correction before the paper's headline claims can be accepted.

major comments (3)
  1. [§3.3, Definitions 6–7 and Proposition 1] Definition 6 counts only additional utility evaluations beyond those required to compute ψ(U) for each U∈U. For any O(1)-size candidate set, Algorithm 2 uses zero additional utility evaluations by construction, so Proposition 1 is true under Definition 6 regardless of the structure of U; it does not demonstrate that manipulation is low-cost relative to an honest evaluator computing one semivalue. If the intended baseline is a single semivalue computation, as §3.3 states, then Algorithm 2 with exact semivalues requires |U|·2^N utility evaluations, which is not polynomial additional work. The gameability definition should be reformulated in end-to-end cost terms, and Proposition 1 should be re-derived under that definition; as written, the low-cost adversarial-strategy claim is forced by the metric rather than demonstrated.
  2. [§D.3 and Algorithm 4] The coefficient used for U(S) in the aggregate favorability is incorrect. From Definition 1, summing ψ_j(U) over j∈P gives coefficient |S∩P|·w_{|S|-1} − (|P|−|S∩P|)·w_{|S|} on U(S). The proof of Proposition 2 instead defines α(k,l) = l·w_{k-1} + (k−l)·w_k, and Algorithm 4 line 14 uses (l·w_k − (k−l)·w_{k−1}); both differ from the correct coefficient in sign and in which weight multiplies the subtractive term. Therefore the quantity optimized by Algorithm 4 is not F_agg(ψ(U);P), and Proposition 2 is not proven as stated.
  3. [§C, Algorithm 1, and Lemma 1] Algorithm 1 computes X^-_{i,k} and X^+_{i,k} as stratum averages (Eq. 3), but line 10 forms ψ̂_i(U,w_k) = Σ_k w_k (X^+_{i,k} − X^-_{i,k}) without the binomial coefficient binom(N−1,k). Definition 1 sums marginal contributions over all subsets, so with average estimates the correct estimator is Σ_k w_k·binom(N−1,k)·(X^+_{i,k} − X^-_{i,k}). Lemma 1's proof implicitly inserts this binomial coefficient, making the proof inconsistent with the pseudocode. Since all experiments and Algorithms 2–5 rely on Algorithm 1, the reported valuation magnitudes and range/payout figures are not reproducible from the pseudocode as written.
minor comments (4)
  1. [§D.4, proof of Proposition 3] The decomposition ψ(U_{p_t}) = ψ(U_T) + (1/(1−p_t))·ψ(U_F) has the wrong sign and the following line uses U_T twice; the correct relation from Definition 5 is ψ(U_{p_t}) = ψ(U_T) − (p_t/(1−p_t))·ψ(U_F). The endpoint conclusion still follows because the coefficient of ψ(U_F) is monotone in p_t, but the displayed equations should be corrected.
  2. [§4.2] The text refers to 'Figure 4.2' where it should refer to Figure 2.
  3. [§3.3] The phrase 'lackspoly(N )complexity sought per 7' should read 'lacks the poly(N) complexity sought per Definition 7'.
  4. [Proposition 2 statement] The statement mixes 'U ⊆ Usmall' with 'This candidate set is U0(k*)'; the quantifiers and notation should be aligned with the proof's construction of U^b.

Circularity Check

1 steps flagged · score 5.0 of 10

Gameability of constant-size utility sets is an artifact of Definition 6/7; the arbitrariness analysis is not circular.

  1. self definitional [Section 3.3, Definition 6; Proposition 1; Algorithm 2]
    "A candidate utility class U is said to be exactly gameable under favorability function F and semivalue ψ if, for any subset P⊆ D, there exists an algorithm that computes U ∗ ∈arg maxU∈U F (ψ(U ); P )using at most O(poly(N ))additional utility function evaluations beyond those required to computeψ(U)for eachU∈ U. ... Let U be a finite candidate set of utility functions of size O(1), i.e., independent of the dataset size |D| = N. U is( ϵ, δ)-gameable under favorabilityF agg(ψ(U);P)for anyPusing Algorithm 2."

    Definition 6 measures gameability as overhead beyond computing ψ(U) for every U in the candidate set. Algorithm 2 does exactly one semivalue evaluation per candidate and then returns the maximizing utility, so for any finite O(1) candidate set the 'additional' cost is zero by construction. Proposition 1 is therefore not an independent demonstration of a low-cost adversarial strategy; it is a restatement of the definition. The paper's own justification—'Given the baseline cost of computing most semivalues is O(2^N), only modest overhead is needed'—shows that the headline 'low-cost' claim discounts the already-exponential per-candidate semivalue computation. The finite-set gameability result reduces to the chosen definition rather than to a nontrivial algorithmic construction.

full rationale

The arbitrariness analysis in Sections 3.2 and 4 is self-contained: the ambiguity sets (Definitions 2-5) are defined from observable modeling choices, and the reported ranges of payout/rank/filter outcomes are computed directly from those candidate sets on benchmark datasets, with no fitted parameter being relabeled as a prediction. The self-citation to Diehl and Wilson (2025) is contextual and not load-bearing for any theorem. The score is elevated only because the gameability contribution is partially circular: Definition 6/7 defines gameability as overhead beyond the semivalue computation, which makes Proposition 1 true by construction for every constant-size candidate set. The other gameability results (Propositions 2 and 3) do use real structure (strata decomposition, linearity of net benefit), but Proposition 1 is the paper's basic 'simple scenario' and the proof of Proposition 3 explicitly invokes it; the notion of 'low-cost' in the abstract thus depends on the definition's decision to discount exponential semivalue cost. A separate correctness issue (the coefficient on U(S) in Algorithm 4 / Appendix D.3 appears inconsistent with Definition 1) is a technical flaw rather than a circularity and does not further change the circularity score.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical or statistical entities are introduced; the free parameters are the ambiguity-set instantiation choices used in the experiments, and the axioms are the semivalue framework plus the boundedness/variance assumptions that make the gameability results tractable.

free parameters (3)
  • small-data threshold kmin = 0.1|D| for experiments
    In Section 4, all coalitions below size 0.1|D| are mapped to U(empty set); the threshold is chosen for illustration and is a degree of freedom the paper itself uses to generate ambiguity.
  • cost-ratio interval [a,b] = [0.5, 0.6] discretized into 100 utilities
    The adversarial candidate set U2 (Def. 5) is instantiated only on this interval; the interval is arbitrary and affects the reported ranges.
  • stratified sampling budgets m_ik = not specified
    Algorithm 1 requires per-cardinality sampling budgets; the paper does not state them, making the empirical ranges unreproducible.
assumptions (4)
  • standard math Semivalue axioms (linearity, anonymity, dummy) define the value allocation from a utility function.
    Used throughout, e.g., Definition 1 and Appendix A, to justify the semivalue operator and the linearity of psi in U.
  • domain assumption Assumption 1: utility candidate values have bounded range r over all coalitions.
    Introduced in Sec. 3.3 to bound stratum variance and provide (epsilon, delta) gameability runtime guarantees.
  • domain assumption Assumption 2: stratum variance sigma^2_{i,k}(U) scales at worst polynomially in N.
    Stated in Appendix D.1; needed for the sampling-based gameability result under Definition 7.
  • ad hoc to paper Gameability definitions count only additional utility evaluations beyond computing one semivalue.
    Definition 6/7 makes Prop. 1 (finite O(1) candidate sets are gameable) true by construction, so the practical significance of gameable depends on accepting this baseline.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Semivalue-based data valuation is arbitrary and gameable." pith.science (2026). https://pith.science/paper/BTZKXUJI

@misc{pith2026250612619,
  author       = {Pith},
  title        = {Pith review of: Semivalue-based data valuation is arbitrary and gameable},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BTZKXUJI}},
  note         = {Machine review of arXiv:2506.12619}
}
read the original abstract

The game-theoretic notion of the semivalue offers a popular framework for credit attribution and data valuation in machine learning. Semivalues have been proposed for a variety of high-stakes decisions involving data, such as determining contributor compensation, acquiring data from external sources, or filtering out low-value datapoints. In these applications, semivalues depend on the specification of a utility function that maps subsets of data to a scalar score. While it is broadly agreed that this utility function arises from a composition of a learning algorithm and a performance metric, its actual instantiation involves numerous subtle modeling choices. We argue that this underspecification leads to varying degrees of arbitrariness in semivalue-based valuations. Small, but arguably reasonable changes to the utility function can induce substantial shifts in valuations across datapoints. Moreover, these valuation methodologies are also often gameable: low-cost adversarial strategies exist to exploit this ambiguity and systematically redistribute value among datapoints. Through theoretical constructions and empirical examples, we demonstrate that a bad-faith valuator can manipulate utility specifications to favor preferred datapoints, and that a good-faith valuator is left without principled guidance to justify any particular specification. These vulnerabilities raise ethical and epistemic concerns about the use of semivalues in several applications. We conclude by highlighting the burden of justification that semivalue-based approaches place on modelers and discuss important considerations for identifying appropriate uses.

Figures

Figures reproduced from arXiv: 2506.12619 by the authors.

Figure 1
Figure 1. Distribution of individual observations’ payout sensitivity to each explored candidate [PITH_FULL_IMAGE:figures/full_fig_p011_1.png] view at source ↗
Figure 4
Figure 4. depicts the variability of the scaled rank favorability outcome for each single-choice [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 2
Figure 2. Comparison of semivalue notions (Shapley vs. Banzhaf) in terms of the distribution of [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figures from the paper (1 more)
Figure 3
Figure 3. Figure 3: Distribution of aggregate payout sensitivity to each explored candidate alternative set (4) [PITH_FULL_IMAGE:figures/full_fig_p019_3.png]

Discussion (0). Sign in 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. An Asymptotic Analysis of the Shapley Value for Dataset Valuation

    cs.GT 2026-07 conditional novelty 7.0 of 10

    Under smooth RKHS embedding utilities, a fixed owner's Shapley value is O(1/I)-close in L1 to an explicit leading term of scale (log I)/I driven by a first-order population signal.

Reference graph

Works this paper leans on

39 extracted references · 21 canonical work pages · cited by 1 Pith paper

  1. [1]

    Agarwal, M

    A. Agarwal, M. Dahleh, and T. Sarkar. A Marketplace for Data : An Algorithmic Solution , May 2019. URL http://arxiv.org/abs/1805.08125

  2. [2]

    Barocas, M

    S. Barocas, M. Hardt, and A. Narayanan. Fairness and Machine Learning: Limitations and Opportunities. MIT Press, 2023

  3. [3]

    Black, M

    E. Black, M. Raghavan, and S. Barocas. Model Multiplicity : Opportunities , Concerns , and Solutions . In 2022 ACM Conference on Fairness , Accountability , and Transparency , pages 850--863, Seoul Republic of Korea, June 2022. ACM. ISBN 978-1-4503-9352-2. doi:10.1145/3531146.3533149. URL https://dl.acm.org/doi/10.1145/3531146.3533149

  4. [4]

    Bornschein, F

    J. Bornschein, F. Visin, and S. Osindero. Small data, big decisions: Model selection in the small-data regime. CoRR, abs/2009.12583, 2020. URL https://arxiv.org/abs/2009.12583

  5. [5]

    Fair Pipelines

    A. Bower, S. N. Kitchen, L. Niss, M. J. Strauss, A. Vargas, and S. Venkatasubramanian. Fair Pipelines , July 2017. URL http://arxiv.org/abs/1707.00391. arXiv:1707.00391 [cs]

  6. [6]

    Cheong, B

    H. Cheong, B. Kim, and I. U. Vaquero. A Data Valuation Model to Estimate the Investment Value of Platform Companies : Based on Discounted Cash Flow . Journal of Risk and Financial Management, 16 0 (6): 0 293, June 2023. ISSN 1911-8074. doi:10.3390/jrfm16060293. URL https://www.mdpi.com/1911-8074/16/6/293. Number: 6 Publisher: Multidisciplinary Digital Pub...

  7. [7]

    Creel and D

    K. Creel and D. Hellman. The Algorithmic Leviathan : Arbitrariness , Fairness , and Opportunity in Algorithmic Decision - Making Systems . Canadian Journal of Philosophy, 52 0 (1): 0 26--43, Jan. 2022. ISSN 0045-5091, 1911-0820. doi:10.1017/can.2022.3. URL https://www.cambridge.org/core/product/identifier/S0045509122000030/type/journal_article

  8. [8]

    Diehl and A

    H. Diehl and A. C. Wilson. The surprising amount of arbitrariness in shapley-value data valuation. Navigating and Addressing Data Problems for Foundation Models (DATA-FM @ ICLR), 2025

Show all 39 references
  1. [9]

    Dwork and C

    C. Dwork and C. Ilvento. Fairness Under Composition . LIPIcs, Volume 124, ITCS 2019, 124: 0 33:1--33:20, 2019. ISSN 1868-8969. doi:10.4230/LIPIcs.ITCS.2019.33. URL http://arxiv.org/abs/1806.06122. arXiv:1806.06122 [cs]

  2. [10]

    Dwork, C

    C. Dwork, C. Ilvento, G. N. Rothblum, and P. Sur. Abstracting Fairness : Oracles , Metrics , and Interpretability , Apr. 2020. URL http://arxiv.org/abs/2004.01840. arXiv:2004.01840 [cs]

  3. [11]

    Enshaei, M

    N. Enshaei, M. J. Rafiee, A. Mohammadi, and F. Naderkhani. Data Shapley Value for Handling Noisy Labels : An application in Screening COVID -19 Pneumonia from Chest CT Scans , Oct. 2021. URL http://arxiv.org/abs/2110.08726. arXiv:2110.08726 [cs, eess]

  4. [12]

    Fabris, S

    A. Fabris, S. Messina, G. Silvello, and G. A. Susto. Algorithmic Fairness Datasets : the Story so Far . Data Mining and Knowledge Discovery, 36 0 (6): 0 2074--2152, Nov. 2022. ISSN 1384-5810, 1573-756X. doi:10.1007/s10618-022-00854-z. URL http://arxiv.org/abs/2202.01711. arXiv...

  5. [13]

    Ganesh, A

    P. Ganesh, A. Taik, and G. Farnadi. The Curious Case of Arbitrariness in Machine Learning , Jan. 2025. URL http://arxiv.org/abs/2501.14959. arXiv:2501.14959 [cs] version: 1

  6. [14]

    Ghorbani and J

    A. Ghorbani and J. Zou. Data Shapley : Equitable Valuation of Data for Machine Learning , June 2019

  7. [15]

    Ghorbani, M

    A. Ghorbani, M. P. Kim, and J. Zou. A Distributional Framework for Data Valuation , Feb. 2020. URL https://arxiv.org/abs/2002.12334v1

  8. [16]

    D. Han, M. Wooldridge, A. Rogers, O. Ohrimenko, and S. Tschiatschek. Replication- Robust Payoff - Allocation for Machine Learning Data Markets . IEEE Transactions on Artificial Intelligence, 4 0 (5): 0 1114--1128, Oct. 2023. ISSN 2691-4581. doi:10.1109/TAI.2022.3195686. URL ht...

  9. [17]

    Hosmer and S

    D. Hosmer and S. Lemeshow. Applied Logistic Regression. John Wiley & Sons, Ltd, 2000. ISBN 9780471722144

  10. [18]

    R. Jia, D. Dao, B. Wang, F. A. Hubis, N. M. G \" u rel, B. Li, C. Zhang, C. J. Spanos, and D. Song. Efficient task-specific data valuation for nearest neighbor algorithms. CoRR, abs/1908.08619, 2019 a . URL http://arxiv.org/abs/1908.08619

  11. [19]

    R. Jia, D. Dao, B. Wang, F. A. Hubis, N. Hynes, N. M. Gurel, B. Li, C. Zhang, D. Song, and C. Spanos. Towards Efficient Data Valuation Based on the Shapley Value , Feb. 2019 b . URL https://arxiv.org/abs/1902.10275v4

  12. [20]

    K. F. Jiang, W. Liang, J. Zou, and Y. Kwon. OpenDataVal : a Unified Benchmark for Data Valuation , June 2023. URL http://arxiv.org/abs/2306.10577. arXiv:2306.10577 [cs, stat]

  13. [21]

    Jurcys, C

    P. Jurcys, C. Donewald, M. Fenwick, M. Lampinen, and A. Smaliukas. Ownership of User - Held Data : Why Property Law Is the Right Approach . SSRN Electronic Journal, 2020. ISSN 1556-5068. doi:10.2139/ssrn.3711017. URL https://www.ssrn.com/abstract=3711017

  14. [22]

    Kleinberg, C

    J. Kleinberg, C. H. Papadimitriou, and P. Raghavan. On the value of private information. In Proceedings of the 8th conference on Theoretical aspects of rationality and knowledge , TARK '01, pages 249--257, San Francisco, CA, USA, July 2001. Morgan Kaufmann Publishers Inc. ISBN...

  15. [23]

    Kumar, A

    S. Kumar, A. Lakshminarayanan, K. Chang, F. Guretno, I. H. Mien, J. Kalpathy-Cramer, P. Krishnaswamy, and P. Singh. Towards More Efficient Data Valuation in Healthcare Federated Learning using Ensembling , Sept. 2022. URL http://arxiv.org/abs/2209.05424. arXiv:2209.05424 [cs]

  16. [24]

    Kwon and J

    Y. Kwon and J. Zou. Beta Shapley : a Unified and Noise -reduced Data Valuation Framework for Machine Learning , Jan. 2022. URL http://arxiv.org/abs/2110.14049. arXiv:2110.14049 [cs, stat]

  17. [25]

    Maleki, L

    S. Maleki, L. Tran-Thanh, G. Hines, T. Rahwan, and A. Rogers. Bounding the Estimation Error of Sampling -based Shapley Value Approximation , Feb. 2014. URL http://arxiv.org/abs/1306.4265. arXiv:1306.4265 [cs]

  18. [26]

    Moody and P

    D. Moody and P. Walsh. Measuring The Value Of Information : An Asset Valuation Approach . In European Conference on Information Systems, 1999. URL https://api.semanticscholar.org/CorpusID:9136893

  19. [27]

    Namba, S

    H. Namba, S. Horiguchi, M. Hamamoto, and M. Egi. Thresholding data shapley for data cleansing using multi-armed bandits, 2024. URL https://arxiv.org/abs/2402.08209

  20. [28]

    Schoch, R

    S. Schoch, R. Mishra, and Y. Ji. Data Selection for Fine -tuning Large Language Models Using Transferred Shapley Values , June 2023. URL http://arxiv.org/abs/2306.10165. arXiv:2306.10165 [cs]

  21. [29]

    L. S. Shapley. A Value for N - Person Games . Technical report, RAND Corporation, Mar. 1952. URL https://www.rand.org/pubs/papers/P295.html

  22. [30]

    Ingredients

    R. H. L. Sim, X. Xu, and B. K. H. Low. Data Valuation in Machine Learning : " Ingredients ", Strategies , and Open Challenges . In Proceedings of the Thirty - First International Joint Conference on Artificial Intelligence , Vienna, Austria, July 2022. International Joint Conf...

  23. [31]

    S. Tang, A. Ghorbani, R. Yamashita, S. Rehman, J. A. Dunnmon, J. Zou, and D. L. Rubin. Data valuation for medical imaging using Shapley value and application to a large-scale chest X -ray dataset. Scientific Reports, 11 0 (1), Apr. 2021. ISSN 2045-2322. doi:10.1038/s41598-021-...

  24. [32]

    Z. Tian, J. Liu, J. Li, X. Cao, R. Jia, J. Kong, M. Liu, and K. Ren. Private Data Valuation and Fair Payment in Data Marketplaces , Feb. 2023. URL http://arxiv.org/abs/2210.08723

  25. [33]

    J. T. Wang and R. Jia. Data Banzhaf : A Robust Data Valuation Framework for Machine Learning , Mar. 2023. URL http://arxiv.org/abs/2205.15466

  26. [34]

    J. T. Wang, Z. Deng, H. Chiba-Okabe, B. Barak, and W. J. Su. An Economic Solution to Copyright Challenges of Generative AI , Sept. 2024. URL http://arxiv.org/abs/2404.13964. arXiv:2404.13964 [cs]

  27. [35]

    T. Wang, J. Rausch, C. Zhang, R. Jia, and D. Song. A Principled Approach to Data Valuation for Federated Learning , Sept. 2020. URL http://arxiv.org/abs/2009.06192. arXiv:2009.06192 [cs, stat]

  28. [36]

    M. Wu, R. Jia, C. Lin, W. Huang, and X. Chang. Variance reduced Shapley value estimation for trustworthy data valuation, May 2023. URL http://arxiv.org/abs/2210.16835

  29. [37]

    Yan and A

    T. Yan and A. D. Procaccia. If You Like Shapley Then You ’ll Love the Core . Proceedings of the AAAI Conference on Artificial Intelligence, 35 0 (6), May 2021. ISSN 2374-3468. doi:10.1609/aaai.v35i6.16721. URL https://ojs.aaai.org/index.php/AAAI/article/view/16721

  30. [38]

    Zhang, Y

    J. Zhang, Y. Bi, M. Cheng, J. Liu, K. Ren, Q. Sun, Y. Wu, Y. Cao, R. C. Fernandez, H. Xu, R. Jia, Y. Kwon, J. Pei, J. T. Wang, H. Xia, L. Xiong, X. Yu, and J. Zou. A Survey on Data Markets , Nov. 2024. URL http://arxiv.org/abs/2411.07267. arXiv:2411.07267 [cs]

  31. [39]

    L. Zhu, H. Dong, M. Shen, and K. Gai. An Incentive Mechanism Using Shapley Value for Blockchain - Based Medical Data Sharing . In 2019 IEEE 5th Intl Conference on Big Data Security on Cloud ( BigDataSecurity ), IEEE Intl Conference on High Performance and Smart Computing , ( H...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.