Pith. sign in

REVIEW 3 major objections 5 minor 41 references

Stochastic SketchRefine: Scaling In-Database Decision-Making under Uncertainty to Millions of Tuples

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Risk-constrained stochastic package queries over uncertain relations can be solved near-optimally at million-tuple scale by combining risk-constraint linearization with a stochastic divide-and-conquer sketch-refine framework.

desk verdict Credible empirical systems paper with an overclaimed theoretical guarantee: the (1−ε)² bound as stated is not proven, but the experiments stand on their own. read the letter →

arxiv 2411.17915 v2 pith:M5GRP3VI submitted 2024-11-26 cs.DB

Riddho R. Haque , Anh L. Mai , Matteo Brucato , Azza Abouzied , Peter J. Haas , Alexandra Meliou This is my paper · ORCID
classification cs.DB MSC 68P1590C1090C15
keywords stochasticpackagequeriesValue-at-RiskConditionalrisk-constraintlinearizationsketch-refineintegerlinearprogramminguncertaindatabasesin-databaseoptimization
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

This paper tries to show that risk-constrained stochastic package queries — choosing a bag of tuples that maximizes an expected objective while keeping tail risks such as Value-at-Risk (VaR) or Conditional Value-at-Risk (CVaR) within limits — can be solved near-optimally at million-tuple scale. The obstacle was that standard Monte Carlo formulations grow with both the number of scenarios and the number of tuples, so prior solvers take hours or fail outright. The paper's two mechanisms are risk-constraint linearization (RCL), which replaces each nonlinear risk constraint with a linearized CVaR constraint, and Stochastic SketchRefine, which partitions the relation into small similar groups, solves a sketch problem over correlated duplicate representatives, and then refines the solution back to actual tuples. Experiments show orders-of-magnitude runtime improvements over the prior state of the art and successful query execution on relations with millions of tuples. If correct, this makes in-database stochastic optimization practical for portfolio selection, procurement, and similar decisions under uncertainty.

What carries the argument

The machinery is the linearized CVaR (L-CVaR) constraint, defined as the sum over package tuples of each tuple's own CVaR times its multiplicity, plus the two-phase Stochastic SketchRefine pipeline built on partitions of similar tuples. The L-CVaR constraint converts each VaR or CVaR risk condition into a single linear inequality, so the SAA ILP's size stops growing with the number of scenarios; the sketch phase solves a small ILP over duplicated representatives of each partition, and the refine phase replaces those duplicates with actual tuples partition by partition. The argument's load-bearing identities are Theorem 2.1 (an L-CVaR constraint is stricter than the corresponding CVaR constraint, which is stricter than the VaR constraint), Theorem 3.1 (the alternating parameter search reaches the optimal L-CVaR parameterization in the idealized setting), and Theorem E.1 (if the optimal package is sketch-feasible, the returned package attains at least $(1-\epsilon)^2$ times the optimal objective minus a diameter-dependent term).

What would settle it

Construct a partition where all tuples have CVaR values systematically above or below their representative's CVaR — for example by adding a common shift within the diameter bound — and run a query whose optimal package uses several tuples from that partition. If Stochastic SketchRefine returns NULL or a package whose objective falls below $(1-\epsilon)^2$ of the relaxed optimal value, then sketch feasibility or Theorem F.1's uniformity assumption has failed. A reader can also simulate the paper's own bound: with $C$ independent constraints at $\delta = 0$, Theorem F.1 gives sketch-feasibility probability $1/2^C$, so a workload with several such constraints should show failures at a measurable rate if the assumption is wrong.

Watch

Extended reading notes

Core claim

The paper's central claim is that stochastic package queries with VaR and CVaR constraints can be evaluated near-optimally at million-tuple scale in minutes by solving integer linear programs whose size does not depend on the number of Monte Carlo scenarios. Risk-constraint linearization searches for L-CVaR parameter values so that the reshaped feasible region contains a validation-feasible and epsilon-optimal package, and Stochastic SketchRefine divides the relation into small similar partitions, solves a sketch problem over correlated duplicate representatives, then refines the solution back to real tuples. The paper reports that the combined method produces packages of quality comparable to the prior state of the art while running orders of magnitude faster, and that it handles relations with up to 4.8 to 6 million tuples where previous methods time out or exhaust memory.

Load-bearing premise

The near-optimality guarantee holds only if the optimal package can be mapped into the sketch without violating feasibility, and the paper's only probability bound for that event assumes that, within each partition, tuple CVaR values are uniformly and independently scattered around the representative's CVaR, an assumption the paper itself calls strong.

Editorial extensions

If this is right

  • Risk-constrained package queries over high-variance relations become practical: workloads that previously took hours or failed are reported to run in minutes.
  • The number of optimization scenarios no longer drives ILP size for VaR/CVaR constraints, so the scenario blowup that plagued prior methods is removed rather than merely compressed.
  • In-database decision-making can now include coherent CVaR risk constraints without a separate optimization stack, because the SPaQL language is extended to express them.
  • Package quality is tunable: the user's error bound $\epsilon$ limits the final objective to at least $(1-\epsilon)^2$ of the relaxed optimum, provided partitions are tight enough for sketch feasibility.
  • The method inherits the deterministic SketchRefine benefit of keeping every ILP subproblem small enough for in-memory solvers, avoiding the memory crashes that limit prior methods on million-tuple relations.

Reading between the lines

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

  • Not claimed in the paper: the same L-CVaR linearization could be lifted out of the database setting and applied to any chance-constrained MILP where VaR/CVaR constraints currently burden the scenario count, so the technique may generalize to supply-chain and energy scheduling problems.
  • Not claimed in the paper: the uniformity and independence assumption in Theorem F.1 is the point most worth stress-testing; an empirical study measuring within-partition CVaR deviation distributions on real uncertain data would show how often the sketch-feasibility condition actually holds.
  • Not claimed in the paper: the refine phase's partition groups are independent once the sketch package is fixed, so the approach should parallelize almost linearly across cores, making the reported minute-scale runtimes a conservative ceiling rather than a floor.
  • Not claimed in the paper: DistPartition's MAD distance, which already accounts for correlation between tuples, could be reused as a general clustering distance for uncertain data well beyond package queries.
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 / 5 minor

Summary. The paper proposes two mechanisms for scaling stochastic package queries (SPQs) over uncertain relations. RCL-Solve replaces VaR/CVaR risk constraints with linearized CVaR (L-CVaR) constraints, so that the resulting SAA/ILP problem size does not grow with the number of Monte Carlo scenarios. Stochastic SketchRefine extends deterministic SketchRefine by partitioning the relation with DistPartition, building correlated duplicates of partition representatives, solving a sketch problem over the duplicates, and then refining the sketch solution back to actual tuples. The paper claims that RCL alone outperforms SummarySearch, that RCL plus Stochastic SketchRefine scales to millions of tuples in minutes, and that Stochastic SketchRefine enjoys a (1-epsilon)^2 approximation guarantee under a sketch-feasibility condition. The experimental section reports 16-run averages on two workloads (portfolio and TPC-H) with up to 6M tuples, ablation studies for duplicates and partitioning, and robustness checks.

Significance. If the formal guarantees were valid, this would be an important step toward practical in-database optimization under uncertainty, since it attacks both the scenario dimension and the tuple dimension simultaneously. The experimental evidence is extensive and reproducible in spirit: 16-run error bars, two datasets, up to 6M tuples, and ablations. The paper also provides a new language construct (L-CVaR) and a partitioning method (DistPartition) with sub-quadratic complexity, which are of independent interest. However, as detailed below, the main approximation theorem is not proven as stated, and the sketch-feasibility probability result rests on an explicitly strong, unverified distributional assumption. The practical claims may survive, but the theoretical contribution needs substantial repair.

major comments (3)
  1. [Appendix E, Theorem E.1] The proof of Theorem E.1 does not establish the stated guarantee. In the proof, after deriving the chain for the mapped package, the text asserts "Σ_i E[t_i.O] x*_i – d_o Σ_i x*_i ≥ ω* – d_o Pmax". This inequality uses Σ_i E[t_i.O] x*_i ≥ ω*, which holds only when x* is an optimal package for Q(S). But the theorem statement quantifies over the package x* returned by Stochastic SketchRefine, and the returned package is not assumed to be optimal for Q(S). Consequently, the derived lower bound ω*_sk ≥ ω* – d_o Pmax has no basis for an arbitrary returned package, and the chain to ω ≥ (1−ε)^2(ω* – d_o Pmax) collapses. The theorem needs either a corrected quantifier over the optimal package (with a separate argument linking the returned package to the optimal package), or a new derivation that uses the objective value of the returned package itself and proves a meaningful comparison to ω*. As written, the headline (1−ε)^2 guarantee is not established.
  2. [Appendix F, Theorem F.1] The only probabilistic support for sketch feasibility, Theorem F.1, relies on assumption (ii): the CVaR values of tuples in each partition are independently and uniformly distributed around the CVaR of the representative. This assumption is not implied by the MAD/diameter conditions used elsewhere in the paper; Theorem 5.1 provides only a deterministic bound |CVaR_α(t1.C) − CVaR_α(t2.C)| ≤ d_C/α, with no uniformity or independence. The paper itself calls the assumption strong, and indeed for C independent constraints the resulting sketch-feasibility probability can be as low as 1/2^C, as the paper notes. No empirical evidence is given that the uniform model holds for the GBM or Gaussian datasets. Therefore the statement in Section 5 that "use of MAD allows us to formally guarantee that Stochastic SketchRefine achieves a (1−ε)^2-optimal solution" is not supported: the guarantee is conditional on an unverified event. The authors should either derive a high-probability bound from the MAD/diameter properties (which does not seem possible without additional assumptions), or explicitly reframe the contribution as an empirical method with a conditional guarantee that is clearly stated.
  3. [Section 3 and Appendix G] The paper's own caveats undermine the unconditional form of the near-optimality claim. Section 3 states that if the alternating-parameter search fails, RCL-Solve "will return the best validation-feasible package encountered so far, but with no optimality guarantees," and that while this behavior is theoretically possible it did not occur in the experiments. Appendix G similarly states that "Stochastic SketchRefine usually provides a similar guarantee in the order of (1±ε)^2". These qualifications are appropriate, but they should be reflected in the abstract and in the formal statement of guarantees. As written, the abstract's claim that solving the linearized ILPs "gives us feasible and near-optimal packages" is stronger than what is proven. The paper should either prove the missing success conditions for the parameter search, or clearly present the guarantee as conditional on the search not failing, with the experimental frequency of success reported.
minor comments (5)
  1. [Appendix C, Algorithm 3] The proposed O(d) transformation for generating correlated normal samples does not appear to produce the stated covariance structure. For example, using the definitions in Algorithm 3, Cov(s'_1, s'_2) evaluates to −λ1 − λ2 rather than the target correlation κ. Please verify the normalization and the eigenvector representation, and either provide a corrected formula or point to the code used in the experiments.
  2. [Appendix F, Theorem F.1 statement vs. proof] The theorem statement uses the bound floor(s/2 − δα/d_C), while the proof derives the Irwin-Hall CDF with argument x = s/2 − δα/(2d_C). The factor of 2 in the denominator is inconsistent between the statement and proof. Please align the expression.
  3. [Section 6.2] The text reports "relative integrality gap in the 0.94–0.99 range" for RCL-Solve, but the metric defined in Section 6.1 is (ω − ω*)/ω*, which is non-positive for any feasible package with ω ≤ ω*. The reported numbers appear to be approximation ratios (ω/ω*), not gaps. Please reconcile the terminology.
  4. [Section 3] The notation for the upper bound on the optimal objective is garbled: the text says "we use an upper bound ω≥ω" and later "above (1−ε)ω". Please use distinct symbols (e.g., ω̄ for the upper bound and ω* for the true optimum) throughout.
  5. [Section 5] The word "embarassingly" should be "embarrassingly". Also, the conservative diameter test 2·d̂_A ≤ d_A is asserted without proof; a brief justification would help readers understand why the factor 2 is safe.

Circularity Check

1 steps flagged · score 6.0 of 10

Theorem E.1's (1−ε)² guarantee is derived only by assuming the returned package is optimal, making the central near-optimality claim circular; the empirical scaling results are independent.

  1. other [Appendix E, proof of Theorem E.1 (Approximation Guarantee)]
    "Σn i=1 E[ti.O] x∗i−do Σn i=1 x∗i ≥ω∗−do Pmax."

    The theorem's hypothesis only says that Stochastic SketchRefine returns a non-NULL package x*; it does not say x* is optimal. The displayed inequality asserts that the objective of the returned package, Σ E[ti.O] x*_i, is at least the true optimum ω*. That is exactly the near-optimality property the theorem is supposed to prove, and it is stronger than the claimed (1−ε)² bound. Without this assertion, the subsequent chain ω*_sk ≥ ω*_s ≥ ω* − d_o Pmax collapses, because a merely feasible returned package can have objective far below ω*. Thus the central approximation guarantee reduces to an unproved optimality assumption on the very object being bounded; the proof is circular at its load-bearing step.

full rationale

The paper's RCL and DistPartition material is largely self-contained: Theorem 2.1 follows from the CVaR definitions and the cited external reference [29, Prop. 6.9]; Theorem 5.1 is proven directly from MAD and CVaR definitions; the APS optimality argument in Theorem 3.1 is a monotonic-search argument over the parameter space. The empirical scalability comparison against SummarySearch is out-of-sample and does not reduce to fitted parameters, and the robustness check in Appendix I.3 uses test scenarios never used during optimization or validation. The one load-bearing circular step is in Appendix E: the proof of the (1−ε)² approximation guarantee obtains its key lower bound by replacing the returned package's objective with the true optimum ω*, which is the very near-optimality claim the theorem must establish. Theorem F.1, offered as support for the sketch-feasibility condition, rests on the paper's own 'strong assumption' that per-tuple CVaR values are independently and uniformly distributed around the representative's CVaR; this is an unverified modeling assumption, but it is not circular. Self-citations to SketchRefine [5] and SummarySearch [7] are used as implementation baselines and building blocks, not as load-bearing support for the new derivations. Because the central theoretical guarantee reduces to an optimality assumption, the circularity score is 6; the empirical scalability and quality claims remain independent and are not themselves circular.

Assumptions & free parameters 10 free parameters · 6 assumptions · 1 invented entities

The central claim rests on a chain of user-chosen hyperparameters, standard CVaR inequalities, and two explicitly stated but unverified modeling assumptions (sketch feasibility and uniform CVaR spread). No physical entities are introduced; the invented entity is an algorithmic device. The theoretical guarantee is conditional and includes an additive term that depends on the diameter threshold of the objective attribute.

free parameters (10)
  • L-CVaR tail parameter alpha' and bound V' per risk constraint = found by alternating parameter search (APS) per query
    Adjusted to make the package validation-feasible and near-optimal; the search is the core of RCL, so the final result depends on these choices.
  • Diameter thresholds d_A = d_price=10, d_gain=100 for Portfolio; d_price=50, d_quantity=5, d_tax=0.05 for TPC-H
    User-chosen to bound within-partition MAD; chosen in experiments to keep partition count in [tau/10, tau/2]; directly controls representative fidelity and runtime.
  • Initial optimization scenario count m = 100
    Hyperparameter; if SAA mismatch is detected, the algorithm doubles m up to |V|.
  • Validation scenario count = 1,000,000
    Set as large as possible to assess feasibility; affects robustness and runtime.
  • Bisection termination threshold delta = 1e-2 in Section 6, 1e-3 mentioned in Section 3
    Stops alpha/V search; the paper is not fully consistent about the value.
  • Approximation error bound epsilon = 0.05
    Used in optimality checks and guarantees; Appendix G reports that setting 0.01 can cause failures.
  • Partition size threshold tau = 100,000
    Upper bound on ILP size for Gurobi; chosen empirically from solver trials.
  • Package size upper bound Pmax = 30
    Assumed known upper bound on distinct tuples in a package; used to set duplicate counts and appears in the approximation bound.
  • Initial risk tolerance Gamma and decrement DeltaGamma = initial >0.99; DeltaGamma=0.03
    Control the number of duplicates in SolveSketch; the initial value was so large that DeltaGamma had no effect in experiments.
  • Duplicate correlation increment Delta rho = 0.1
    Used to increase inter-duplicate correlation when refine fails; not derived from data.
assumptions (6)
  • standard math CVaR_lower(alpha) of a sum of gains is at least the sum of individual CVaR_lower(alpha) values, per Theorem 2.1 first implication from [29, Prop. 6.9].
    Used to justify that L-CVaR constraints are conservative replacements for VaR/CVaR constraints.
  • domain assumption All stochastic attributes have continuous distributions, so CVaR_alpha(A) = E[A | A <= q_alpha(A)].
    Stated in Section 2 for simplicity; violated by the paper's own discrete Bernoulli example, though footnote 1 gives a correction.
  • ad hoc to paper The optimal package x* is sketch feasible, i.e., there exists a mapping from each tuple to a duplicate of its partition representative that yields a feasible sketch package.
    Needed for Theorem E.1; probability only bounded under the strong uniform-CVaR assumption in Theorem F.1.
  • ad hoc to paper For each partition, the CVaR values of the tuples with respect to a constraint attribute are independently and uniformly distributed around the CVaR of the representative.
    Assumption (ii) of Theorem F.1, described by the paper itself as strong; it yields the 1/2^C sketch-feasibility lower bound for C constraints.
  • standard math MAD satisfies the triangle inequality and bounds CVaR differences, per Theorem 5.1.
    Used by DistPartition's PivotScan to enforce intra-partition similarity without pairwise scans.
  • domain assumption Duplicates of a representative are stochastically identical and correlated with pairwise correlation equal to the median Pearson correlation of the representative with partition tuples, generated via NORTA.
    Design assumption; if the median correlation underestimates within-partition tail risk, refine may fail or return suboptimal packages.
invented entities (1)
  • Stochastically identical correlated duplicates of representatives
    purpose: Enable the sketch ILP to allow multiple tuples from the same partition without overestimating risk, while preserving approximate within-partition correlation.
    Synthetic tuples internal to Stochastic SketchRefine; their adequacy is assessed via ablation in Appendix I.1, but they are not independently observable entities.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Stochastic SketchRefine: Scaling In-Database Decision-Making under Uncertainty to Millions of Tuples." pith.science (2026). https://pith.science/paper/M5GRP3VI

@misc{pith2026241117915,
  author       = {Pith},
  title        = {Pith review of: Stochastic SketchRefine: Scaling In-Database Decision-Making under Uncertainty to Millions of Tuples},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M5GRP3VI}},
  note         = {Machine review of arXiv:2411.17915}
}
read the original abstract

Decision making under uncertainty often requires choosing packages, or bags of tuples, that collectively optimize expected outcomes while limiting risks. Processing Stochastic Package Queries (SPQs) involves solving very large optimization problems on uncertain data. Monte Carlo methods create numerous scenarios, or sample realizations of the stochastic attributes of all the tuples, and generate packages with optimal objective values across these scenarios. The number of scenarios needed for accurate approximation - and hence the size of the optimization problem when using prior methods - increases with variance in the data, and the search space of the optimization problem increases exponentially with the number of tuples in the relation. Existing solvers take hours to process SPQs on large relations containing stochastic attributes with high variance. Besides enriching the SPaQL language to capture a broader class of risk specifications, we make two fundamental contributions towards scalable SPQ processing. First, to handle high variance, we propose risk-constraint linearization (RCL), which converts SPQs into Integer Linear Programs (ILPs) whose size is independent of the number of scenarios used. Solving these ILPs gives us feasible and near-optimal packages. Second, we propose Stochastic SketchRefine, a divide and conquer framework that breaks down a large stochastic optimization problem into subproblems involving smaller subsets of tuples. Our experiments show that, together, RCL and Stochastic SketchRefine produce high-quality packages in orders of magnitude lower runtime than the state of the art.

Figures

Figures reproduced from arXiv: 2411.17915 by the authors.

Figure 1
Figure 1. The gain in the Stock_Investments table is an un [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Solving a stochastic package query (SPQ) with [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Increasing variance or volatility coefficients increases tuple uncertainty as well as query hardness. ( [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: RCL-Solve continues to outperform SummarySearch but fails to scale beyond 1M tuples. The absence of data points indicates that no solutions were found within 1.5 hours. Stochastic SketchRefine scales well as data size and query hardness (𝐻) increase. Each plot shows th…
Figure 6
Figure 6. Figure 6: Consistent with the notation in [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]
Figure 7
Figure 7. Figure 7: Trade-off between values of 𝜇 and 𝜎 2 when main￾taining a fixed value of 𝜇𝐷 = 𝑑 + 𝜖 (𝑑 = 4 and 𝜖 = 0.5) [PITH_FULL_IMAGE:figures/full_fig_p018_7.png]
Figure 8
Figure 8. Figure 8: Minimum required sample size as a function of [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]
Figure 9
Figure 9. Figure 9: Offline pre-computation runtime, which includes representative selection, is minimally higher with [PITH_FULL_IMAGE:figures/full_fig_p025_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 33 canonical work pages

  1. [1]

    https://www.kaggle.com/datasets/paultimothymooney/stock-market- data/data

    [n.d.]. https://www.kaggle.com/datasets/paultimothymooney/stock-market- data/data

    work page

  2. [2]

    Hervé Abdi and Lynne J Williams. 2010. Principal component analysis. Wiley interdisciplinary reviews: computational statistics 2, 4 (2010), 433–459

    2010

  3. [3]

    Shabbir Ahmed and Alexander Shapiro. 2008. Solving chance-constrained sto- chastic programs via sampling and integer programming. In State-of-the-Art Decision-making Tools in the Information-intensive Age . (INFORMS), 261–269

    work page 2008

  4. [4]

    Gordon J Alexander and Alexandre M Baptista. 2004. A comparison of VaR and CVaR constraints on portfolio selection with the mean-variance model. Management science 50, 9 (2004), 1261–1273

    work page 2004

  5. [5]

    Matteo Brucato, Azza Abouzied, and Alexandra Meliou. 2018. Package queries: efficient and scalable computation of high-order constraints. The VLDB Journal 27 (2018), 693–718

    work page 2018

  6. [6]

    Matteo Brucato, Juan Felipe Beltran, Azza Abouzied, and Alexandra Meliou

    work page

  7. [7]

    Haas, and Alexandra Meliou

    Matteo Brucato, Nishant Yadav, Azza Abouzied, Peter J. Haas, and Alexandra Meliou. 2020. Stochastic Package Queries in Probabilistic Databases. In Proc. ACM SIGMOD. 269–283. https://doi.org/10.1145/3318464.3389765

    arXiv 2020

  8. [8]

    Stochastic Package Queries in Probabilistic Databases

    Matteo Brucato, Nishant Yadav, Azza Abouzied, Peter J. Haas, and Alexandra Meliou. 2021. Scalable package queries in relational database systems: Extended version. arXiv preprint arXiv:2103.06784 (2021)

    work page Pith review arXiv 2021

Show all 41 references
  1. [9]

    Sébastien Bubeck. 2015. Convex optimization: Algorithms and complexity. Foun- dations and Trends® in Machine Learning 8, 3-4 (2015), 231–357

    2015

  2. [10]

    Felix Campbell, Bahareh Arab, and Boris Glavic. 2022. Efficient Answering of Historical What-if Queries. In Proceedings of the 48th International Conference on Management of Data (SIGMOD) . 1556–1569. https://doi.org/10.1145/3514221. 3526138

    2022 doi

  3. [11]

    Marco C Campi and Simone Garatti. 2011. A sampling-and-discarding approach to chance-constrained optimization: feasibility and optimality. Journal of opti- mization theory and applications 148, 2 (2011), 257–280

    2011

  4. [12]

    Marco C Campi, Simone Garatti, and Maria Prandini. 2009. The scenario approach for systems and control design. Annual Reviews in Control 33, 2 (2009), 149–157

    2009

  5. [13]

    Siu On Chan, Ilias Diakonikolas, Rocco A Servedio, and Xiaorui Sun. 2014. Near- optimal density estimation in near-linear time using variable-width histograms. Advances in neural information processing systems 27 (2014)

    2014

  6. [14]

    Michael Chau, Reynold Cheng, Ben Kao, and Jackey Ng. 2006. Uncertain data mining: An example in clustering location data. In Advances in Knowledge Dis- covery and Data Mining: 10th Pacific-Asia Conference, PAKDD 2006, Singapore, April 9-12, 2006. Proceedings 10 . Springer, 199–204

    2006

  7. [15]

    Nilesh Dalvi and Dan Suciu. 2007. Efficient query evaluation on probabilistic databases. The VLDB Journal 16 (2007), 523–544

    2007

  8. [16]

    1979.Computers and intractability

    Michael R Garey and David S Johnson. 1979.Computers and intractability. Vol. 174. freeman San Francisco

    1979

  9. [17]

    James E Gentle. 2009. Computational Statistics. Springer

    2009

  10. [18]

    Soumyadip Ghosh and Shane G Henderson. 2003. Behavior of the NORTA method for correlated random vector generation as the dimension increases. ACM Transactions on Modeling and Computer Simulation (TOMACS) 13, 3 (2003), 276–294

    2003

  11. [19]

    Francesco Gullo, Giovanni Ponti, and Andrea Tagarelli. 2008. Clustering un- certain data via k-medoids. In International Conference on Scalable Uncertainty Management. Springer, 229–242

    2008

  12. [20]

    Francesco Gullo, Giovanni Ponti, and Andrea Tagarelli. 2013. Minimizing the variance of cluster mixture models for clustering uncertain objects. Statistical Analysis and Data Mining: The ASA Data Science Journal 6, 2 (2013), 116–135

    2013

  13. [21]

    Francesco Gullo, Giovanni Ponti, Andrea Tagarelli, and Sergio Greco. 2008. A hierarchical algorithm for clustering uncertain data via an information-theoretic approach. In 2008 Eighth IEEE International Conference on Data Mining . IEEE, 821–826

    2008

  14. [22]

    Francesco Gullo, Giovanni Ponti, Andrea Tagarelli, and Sergio Greco. 2017. An information-theoretic approach to hierarchical clustering of uncertain data. In- formation sciences 402 (2017), 199–215

    2017

  15. [23]

    Francesco Gullo and Andrea Tagarelli. 2012. Uncertain centroid based partitional clustering of uncertain data. arXiv preprint arXiv:1203.6401 (2012)

    2012 arXiv

  16. [24]

    Tito Homem-de Mello and Güzin Bayraksan. 2014. Monte Carlo sampling- based methods for stochastic optimization. Surveys in Operations Research and Management Science 19, 1 (2014), 56–85

    2014

  17. [25]

    Shudong Huang, Zhao Kang, Zenglin Xu, and Quanhui Liu. 2021. Robust deep k-means: An effective and simple method for data clustering. Pattern Recognition 117 (2021), 107996

    2021

  18. [26]

    Ravi Jampani, Fei Xu, Mingxi Wu, Luis Perez, Chris Jermaine, and Peter J. Haas

  19. [27]

    Sujin Kim, Raghu Pasupathy, and Shane G Henderson. 2015. A guide to sample average approximation. In Handbook of Simulation Optimization . Springer, 207– 243

    2015

  20. [28]

    Mai, Pengyu Wang, Azza Abouzied, Matteo Brucato, Peter J

    Anh L. Mai, Pengyu Wang, Azza Abouzied, Matteo Brucato, Peter J. Haas, and Alexandra Meliou. 2024. Scaling Package Queries to a Billion Tuples via Hier- archical Partitioning and Customized Optimization. Proc. VLDB Endow. 17, 5 (2024), 1146–1158

    2024

  21. [29]

    McNeil, Rüdiger Frey, and Paul Embrechts

    Alexander J. McNeil, Rüdiger Frey, and Paul Embrechts. 2015. Quantitative Risk Management: Concepts, Techniques and Tools. Princeton University Press

    2015

  22. [30]

    Victor M Panaretos and Yoav Zemel. 2019. Statistical aspects of Wasserstein distances. Annual review of statistics and its application 6, 1 (2019), 405–431

    2019

  23. [31]

    Meikel Poess and Chris Floyd. 2000. New TPC benchmarks for decision support and web commerce. ACM Sigmod Record 29, 4 (2000), 64–71

    2000

  24. [32]

    Krishna Reddy and Vaughan Clinton. 2016. Simulating stock prices using geo- metric Brownian motion: Evidence from Australian companies. Australasian Accounting, Business and Finance Journal 10, 3 (2016), 23–47

    2016

  25. [33]

    Sergey Sarykalin, Gaia Serraino, and Stan Uryasev. 2008. Value-at-Risk vs. Conditional Value-at-Risk in Risk Management and Optimization. InState-of-the- Art Decision-Making Tools in the Information-Intensive Age . INFORMS TutORials in Operations Research, 270—-294. https://do...

    2008

  26. [34]

    Jonathon Shlens. 2014. Notes on Kullback-Leibler divergence and likelihood. arXiv preprint arXiv:1404.2000 (2014)

    2014 arXiv

  27. [35]

    Laurynas Šikšnys and Torben Bach Pedersen. 2016. SolveDB: Integrating opti- mization problem solvers into SQL databases. In Proceedings of the 28th Interna- tional Conference on Scientific and Statistical Database Management . 1–12

    2016

  28. [36]

    Laurynas Siksnys, Torben Bach Pedersen, Thomas Dyhre Nielsen, and Davide Frazzetto. 2021. SolveDB+: Sql-based prescriptive analytics. In Advances in Data- base Technology-24th International Conference on Extending Database Technology, EDBT 2021. OpenProceedings. org, 133–144

    2021

  29. [37]

    Eric K Tokuda, Cesar H Comin, and Luciano da F Costa. 2022. Revisiting agglom- erative clustering. Physica A: Statistical mechanics and its applications 585 (2022), 126433

    2022

  30. [38]

    Kyoung-Gu Woo, Jeong-Hoon Lee, Myoung-Ho Kim, and Yoon-Joon Lee. 2004. FINDIT: a fast and intelligent subspace clustering algorithm using dimension voting. Information and Software Technology 46, 4 (2004), 255–271

    2004

  31. [39]

    folded normal

    Xianchao Zhang, Han Liu, and Xiaotong Zhang. 2017. Novel density-based and hierarchical density-based clustering algorithms for uncertain data. Neural networks 93 (2017), 240–255. 14 A FREQUENTLY USED NOTATIONS For the reader’s convenience, Table 1 displays a list of frequentl...

    2017

  32. [2011]

    ACM Trans

    The Monte Carlo database system: Stochastic analysis close to the data. ACM Trans. Database Syst. 36, 3, Article 18 (2011), 41 pages. https://doi.org/10. 1145/2000824.2000828

    2011

  33. [2015]

    arXiv preprint arXiv:1512.03564 (2015)

    Scalable package queries in relational database systems. arXiv preprint arXiv:1512.03564 (2015)

    2015 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.