Pith. sign in

REVIEW 3 major objections 5 minor 47 references

Prescribe-then-Select: Adaptive Policy Selection for Contextual Stochastic Optimization

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read This paper claims that training a cross-validated tree ensemble to route each context to one of several feasible policies beats the best single policy when no one policy dominates, and matches it otherwise.

desk verdict Useful modular framework for context-dependent policy selection in CSO, but the headline claim rests on an unexamined fold/refit stability assumption and synthetic benchmarks only. read the letter →

arxiv 2509.08194 v2 pith:EJ2ADGJK submitted 2025-09-09 cs.LG stat.ML

Caio de Prospero Iglesias , Kimberly Villalobos Carballo , Dimitris Bertsimas This is my paper · ORCID
classification cs.LGstat.ML
keywords contextualstochasticoptimizationpolicyselectionoptimaltreesmeta-policynewsvendorproblemshipmentplanningcross-validationprescriptiveanalytics
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 claims that in contextual stochastic optimization, where no single decision policy dominates across all covariate regions, a meta-policy that learns which candidate policy to deploy for each observed context can outperform the best single policy overall. To do this it proposes Prescribe-then-Select: first build a library of feasible policies, then train an ensemble of optimal policy trees on cross-validated cost tables to route each context to the policy with the lowest realized cost. The paper proves a small lemma showing that if a region exists where one policy beats the globally best policy by a margin delta, then a policy tree that assigns that region to the better policy gains at least delta times the region's probability. On synthetic newsvendor and shipment-planning benchmarks with designed demand regimes, PS reports statistically significant gains over all single policies in heterogeneous regimes, and it matches the dominant policy when one emerges. The practical point is that decision-makers no longer need to commit to one modeling paradigm in advance.

What carries the argument

Optimal Policy Trees are the selection engine: depth-constrained decision trees trained to minimize empirical cost, not prediction error, over cross-validation cost tables C^(k)_{i,m} = c(pi_m(x_i), y_i). The cost table connects policy prescriptions to actual outcomes, letting the tree learn which policy wins where. The ensemble is formed from K folds times R random-seed repeats, and the final meta-policy is a majority vote over the trees. Lemma 1 is the formal anchor: a policy tree that assigns a region R to a policy with delta advantage over the globally best policy improves expected cost by at least delta times Pr(X in R).

What would settle it

Run the pipeline on a synthetic dataset with a known regime boundary and compare PS against an oracle that routes each test context to the truly lowest-conditional-cost policy: if PS's majority-vote ensemble does not reproduce the boundary, or its average cost exceeds the best single policy when fold-trained and full-trained rankings are made to disagree, the central claim would be falsified.

Watch

Extended reading notes

Core claim

Prescribe-then-Select treats policy selection as a supervised learning problem: for each held-out fold, every candidate policy is trained on the complement fold and evaluated on the held-in contexts with their realized outcomes, producing a cost table; Optimal Policy Trees are trained on these tables to partition the covariate space into regions, each assigned one policy index. At deployment the ensemble votes by majority, and the selected policy, refit on the full training set, makes the decision. Because each candidate policy is feasible by construction, the selected policy remains feasible even under hard constraints. Lemma 1 gives the core guarantee: if a covariate region R has a policy

Load-bearing premise

The load-bearing premise is that policy rankings measured on held-out folds with a single realized outcome per context stay the same after each policy is refit on the full training set; if that ranking flips, the meta-policy will systematically pick a suboptimal candidate.

Editorial extensions

If this is right

  • A decision-maker can keep a library of candidate policies from different paradigms and let data, not a priori judgment, decide which to deploy for each context.
  • In heterogeneous regimes, the meta-policy can beat every single candidate on average, so policies that are weak overall remain useful in the regions where they excel.
  • In homogeneous regimes, the method degrades gracefully: it converges to the dominant policy and incurs negligible regret relative to it.
  • Hard feasibility constraints are preserved because the selector chooses whole feasible policies rather than averaging their decisions.
  • The pipeline is modular: new candidate policies can be added without redesigning the selection step, since the cost table and tree training adapt automatically.

Reading between the lines

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

  • Editorial inference: the cost-table construction silently assumes that policy rankings measured on cross-validation folds survive refitting on the full training set; a smoothed or repeated-outcome cost table could reduce this risk, but the paper does not test it.
  • Editorial inference: the trained trees appear to recover interpretable regime structure such as holidays, months, and early-month contracts, which could serve as a diagnostic of where each policy wins; the paper illustrates this but does not develop it as a tool.
  • Editorial inference: nothing in the experiments prevents the framework from carrying over to other feasible policy families, multi-stage settings, or high-dimensional real-world covariates, but that transfer is unverified.
  • Editorial inference: a direct comparison against an oracle selector with true conditional costs would quantify how much of the reported gain is lost to imperfect tree learning; the paper does not report such a bound.
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 Prescribe-then-Select (PS), a two-stage framework for contextual stochastic optimization. In the first stage, a library of feasible candidate policies is constructed (SAA, point-prediction kNN/RF, and predictive-prescriptive kNN/RF). In the second stage, a meta-policy is learned from cross-validated realized-cost tables using ensembles of Optimal Policy Trees (OPTs); at inference the OPT ensemble votes on a policy for each covariate, and the selected (refit) policy prescribes the decision. Lemma 1 states that if a region R of the covariate space exists where some policy m beats the globally best policy m* by δ in conditional expected cost, then a policy tree assigning m to R and m* elsewhere attains an improvement of at least δ·Pr(X∈R). Experiments on synthetic multi-product newsvendor and two-stage shipment-planning problems compare PS against individual policies across nine training sizes and report statistically significant gains in heterogeneous regimes and convergence to the dominant policy when one emerges.

Significance. If the claims hold, PS is a practically appealing modular meta-policy: it reuses existing prescriptive models, preserves hard feasibility by selecting rather than averaging decisions, and is entirely data-driven with out-of-fold cost estimation and reproducible code. The paper gives explicit credit for a clean cross-validation design and for providing code. However, the evidence is entirely synthetic and the data-generation process is engineered to produce exactly the regime heterogeneity PS exploits, so the generalizability claim rests on plausibility rather than demonstration on real data. The theoretical anchor (Lemma 1) is a near-tautological oracle statement and does not by itself secure the finite-sample, refit-based claims. The main risk is the training/deployment mismatch described below; this is fixable with additional diagnostics, but it currently leaves the central claim not fully supported.

major comments (3)
  1. [§4.3, Algorithm 1; §4.2, Lemma 1] The selector is trained on cost-table entries C^{(k)}_{i,m} = c(π_m^{(-k)}(x_i), y_i), where each π_m is fit on the fold complement I^{(-k)} and y_i is a single realized outcome. At deployment, π_1,...,π_M are refit on the full training set and the OPT ensemble is applied to those refit policies. Lemma 1 is a population statement about conditional expected costs with fixed policies; it does not justify that the empirical ordering learned from fold-trained, single-realization costs is stable when policies are refit. If rankings flip between fold-trained and full-trained policies, the meta-policy can systematically route contexts to suboptimal policies. The paper reports no diagnostic for this stability (e.g., agreement between fold-based and refit-based policy choices, or a nested-CV variant with full refit). This is load-bearing for the abstract's 'consistently outperforms' and 'converge
  2. [§4.2, Lemma 1] The statement and proof assume that an arbitrary region R can be assigned to policy m while its complement is assigned to m*. The implemented OPT, however, partitions the feature space into axis-aligned rectangles, so an arbitrary R is not representable. As written, the lemma overstates the support for the actual algorithm. It should either restrict R to a union of leaf-representable regions or be framed as a guarantee under an oracle over all partitions. This is a formal gap in the paper's mathematical anchor.
  3. [§5.3–5.4.2, Figures 5a–5b] The 'best single policy' baseline is not formally defined in the evaluation protocol. If it is the candidate with the highest average test profit (i.e., selected with test-set hindsight), then the comparison is to an infeasible oracle baseline, and the paper should also report a feasible single-policy selection baseline (e.g., the policy with the best cross-validated cost on the training set). If it is instead a per-sample or per-segment oracle, the definition should be stated. The current presentation makes it hard to interpret the magnitude of the reported gains relative to a practical alternative.
minor comments (5)
  1. [§4.2] 'By construction, the optimal policy tree cannot perform worse than the best single policy' should be qualified: this holds for the exact optimum of Eq. (5), not necessarily for the heuristic output of the OPT algorithm with a limited tree class and regularization.
  2. [§4.3, Lemma 1 proof] Typo: 'it’s complement' should be 'its complement'. Also 'These procedure is explained' in §4.3 should be 'This procedure is explained'.
  3. [§5.3] The OPT hyperparameters D_max, n_min, and λ are listed in Algorithm 1 but their values and tuning procedure (if any) are not reported for the experiments.
  4. [§5.4] Statistical significance is inferred from overlapping or non-overlapping 95% confidence intervals. This is informal; a paired test across the 100 training samples (e.g., paired t-test or Wilcoxon) would be more appropriate for the claim of significant improvement.
  5. [Figures 5a–5b] The figure captions say 'Point indicate averages' — should be 'Points indicate averages'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the meta-policy is trained on out-of-fold realized costs and evaluated on a held-out test horizon; Lemma 1 is a population-level mathematical statement, and the self-citations are background only.

full rationale

The paper's derivation chain is self-contained. PS constructs cost tables C^{(k)}_{i,m}=c(pi_m^{(-k)}(x_i), y_i) from fold-trained policies and held-out outcomes (Section 4.3, Algorithm 1), trains OPT ensembles on those tables, then evaluates the resulting majority-vote selector on a separate test set (Section 5.3). The meta-policy is never fit to test data; the reported gains are out-of-sample. Lemma 1 is a conditional-expectation inequality showing that a tree assigning a region to a better policy improves performance by delta·Pr(X in R); it is a mathematical statement, not an empirical prediction, and does not assume the algorithm finds that tree. The only self-citations (e.g., Bertsimas & Carballo 2023; Bertsimas & Dunn 2019; Bertsimas & Kallus 2020) are standard background for the candidate policy families and are not used to justify the central empirical claim. The known fold/refit ranking-stability issue is a potential correctness risk, not a circularity: even if the selector's rankings were unstable, that would be a failure of generalization, not an equivalence between inputs and outputs.

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

The method adds no new physical entities. The free parameters are mostly experimental design choices: the synthetic data regimes are hand-set to create the heterogeneity the method exploits, and the OPT hyperparameters are undisclosed. The core method itself has few fitted constants; k, B, K, R are standard choices, but they are still free parameters that affect the empirical results.

free parameters (5)
  • k (neighbors in kNN) = 5
    Number of neighbors used for PPt-kNN and PP-kNN; fixed without reported tuning (Section 5.3).
  • B (trees in random forest) = 5
    Number of trees in PPt-RF and PP-RF; fixed in Section 5.3.
  • OPT hyperparameters (D_max, n_min, lambda) = not reported
    Depth, leaf size, and regularization of Optimal Policy Trees are listed in Algorithm 1 but values are not given in the paper; the central experimental results depend on them.
  • K and R (CV folds and repetitions) = 5 and 10
    Ensemble size choice (50 trees); affects variance of the meta-policy.
  • Data generation parameters (baseline B=30, noise scales, regime definitions) = Appendix B/C
    Hand-set to create demand regimes with the exact heterogeneity that the method is designed to exploit; these choices do not come from any external data source.
assumptions (4)
  • domain assumption Best-policy regions are learnable by axis-aligned decision trees
    The OPT selector succeeds only if the conditional best-policy regions are well approximated by axis-aligned rectangles (Section 4.2, Eq. 5).
  • domain assumption Optimal Policy Trees heuristic finds near-optimal solutions to Eq. 5
    The paper adopts the IAI implementation of Amram et al. (2022) without optimality guarantees; the empirical claim depends on the heuristic succeeding (Section 4.2).
  • domain assumption Synthetic data generation is a valid proxy for real CSO heterogeneity
    All performance claims are evaluated on simulated demand models in Appendix B/C; there is no real-world validation.
  • standard math Standard probability laws in Lemma 1 proof
    The proof of the improvement bound uses conditional expectation and total probability; these are standard.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Prescribe-then-Select: Adaptive Policy Selection for Contextual Stochastic Optimization." pith.science (2026). https://pith.science/paper/EJ2ADGJK

@misc{pith2026250908194,
  author       = {Pith},
  title        = {Pith review of: Prescribe-then-Select: Adaptive Policy Selection for Contextual Stochastic Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EJ2ADGJK}},
  note         = {Machine review of arXiv:2509.08194}
}
read the original abstract

We address the problem of policy selection in contextual stochastic optimization (CSO), where covariates are available as contextual information and decisions must satisfy hard feasibility constraints. In many CSO settings, multiple candidate policies--arising from different modeling paradigms--exhibit heterogeneous performance across the covariate space, with no single policy uniformly dominating. We propose Prescribe-then-Select (PS), a modular framework that first constructs a library of feasible candidate policies and then learns a meta-policy to select the best policy for the observed covariates. We implement the meta-policy using ensembles of Optimal Policy Trees trained via cross-validation on the training set, making policy choice entirely data-driven. Across two benchmark CSO problems--single-stage newsvendor and two-stage shipment planning--PS consistently outperforms the best single policy in heterogeneous regimes of the covariate space and converges to the dominant policy when such heterogeneity is absent. All the code to reproduce the results can be found at https://anonymous.4open.science/r/Prescribe-then-Select-TMLR.

Figures

Figures reproduced from arXiv: 2509.08194 by the authors.

Figure 1
Figure 1. End-to-end pipeline. Left: K-fold, R-replicate training produces an ensemble of OPTs, each trained on a distinct held-out fold using its corresponding cost table. We also train each policy on the full dataset to maximize downstream performance. Right: At inference, a new context is routed through the OPT ensemble, and the majority-vote policy is applied via the refit models to produce the final decision. Training ph… view at source ↗
Figure 2
Figure 2. Detailed view of one fold in the training phase. In fold k, models trained on I (−k) produce sample weights or point predictions, mapped to decisions π m(xi) for each i ∈ I(k) and m ∈ ΠM. Evaluating these using the true outcomes yi yields the cost matrix C (k) , which, together with {xi : i ∈ I(k)}, trains R Optimal Policy Trees. Across all folds, this produces R × K selectors. We show an example with M = 3 policies… view at source ↗
Figure 3
Figure 3. Newsvendor: segment-wise profit vs. training size (average and 95% CIs shown). [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Shipment: segment-wise mean test profit vs. training size (95% CIs; profit [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Profit vs. training size for PS and the best single candidate policy in the (left) newsvendor and [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Illustrative OPT selectors for the multi–product newsvendor. Two randomly chosen trees from [PITH_FULL_IMAGE:figures/full_fig_p020_6.png]
Figure 7
Figure 7. Figure 7: Illustrative OPT selectors for the shipment–planning task. Two randomly chosen trees from the [PITH_FULL_IMAGE:figures/full_fig_p020_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 36 canonical work pages

  1. [1]

    Optimal policy trees

    Maxime Amram, Jack Dunn, and Ying Daisy Zhuo. Optimal policy trees. Machine Learning, 111 0 (7): 0 2741--2768, 2022. doi:10.1007/s10994-022-06128-5

    work page doi:10.1007/s10994-022-06128-5 2022

  2. [2]

    The big data newsvendor: Practical insights from machine learning

    Gah-Yi Ban and Cynthia Rudin. The big data newsvendor: Practical insights from machine learning. Operations Research, 67 0 (1): 0 90--108, 2019

    work page 2019

  3. [3]

    Generalization bounds for regularized portfolio selection with market side information

    Thierry Bazier-Matte and Erick Delage. Generalization bounds for regularized portfolio selection with market side information. INFOR: Information Systems and Operational Research, 58 0 (2): 0 374--401, 2020

    work page 2020

  4. [4]

    Robust optimization

    Aharon Ben-Tal, Arkadi Nemirovski, and Laurent El Ghaoui. Robust optimization. 2009

    work page 2009

  5. [5]

    Using a financial training criterion rather than a prediction criterion

    Yoshua Bengio. Using a financial training criterion rather than a prediction criterion. International journal of neural systems, 8 0 (04): 0 433--443, 1997

    work page 1997

  6. [6]

    Bertsimas and J

    D. Bertsimas and J. Dunn. Machine Learning Under a Modern Optimization Lens. Dynamic Ideas LLC, 2019. ISBN 9781733788502. URL https://books.google.com/books?id=g3ZWygEACAAJ

    work page 2019

  7. [7]

    Multistage Stochastic Optimization via Kernels

    Dimitris Bertsimas and Kimberly Villalobos Carballo. Multistage stochastic optimization via kernels. arXiv preprint arXiv:2303.06515, 2023

    work page Pith review arXiv 2023

  8. [8]

    From predictive to prescriptive analytics

    Dimitris Bertsimas and Nathan Kallus. From predictive to prescriptive analytics. Management Science, 66 0 (3): 0 1025--1044, 2020. doi:10.1287/mnsc.2018.3253

    arXiv 2020

Show all 47 references
  1. [9]

    Data-driven optimization: A reproducing kernel hilbert space approach

    Dimitris Bertsimas and Nihal Koduri. Data-driven optimization: A reproducing kernel hilbert space approach. Operations Research, 70 0 (1): 0 454--471, 2022

    2022

  2. [10]

    Data-driven robust optimization

    Dimitris Bertsimas, Vishal Gupta, and Nathan Kallus. Data-driven robust optimization. Mathematical Programming, 167 0 (2): 0 235--292, 2018 a

    2018

  3. [11]

    Robust sample average approximation

    Dimitris Bertsimas, Vishal Gupta, and Nathan Kallus. Robust sample average approximation. Mathematical Programming, 171 0 (1): 0 217--282, 2018 b

    2018

  4. [12]

    Birge and François Louveaux

    John R. Birge and François Louveaux. Introduction to Stochastic Programming. Springer, New York, 2nd edition, 2011. ISBN 978-1-4614-0236-8

    2011

  5. [13]

    Random forests

    Leo Breiman. Random forests. Machine learning, 45 0 (1): 0 5--32, 2001

    2001

  6. [14]

    On distributionally robust chance-constrained linear programs

    Giuseppe Carlo Calafiore and L El Ghaoui. On distributionally robust chance-constrained linear programs. Journal of Optimization Theory and Applications, 130 0 (1): 0 1--22, 2006

    2006

  7. [15]

    Decision making under uncertainty in electricity markets, volume 1

    Antonio J Conejo, Miguel Carri \'o n, Juan M Morales, et al. Decision making under uncertainty in electricity markets, volume 1. Springer, 2010

    2010

  8. [16]

    Cover and P

    T. Cover and P. Hart. Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13 0 (1): 0 21--27, 1967. doi:10.1109/TIT.1967.1053964

    1967

  9. [17]

    Hall, Yun Shi, and Tianyuan Su

    Xiangyu Cui, Nicholas G. Hall, Yun Shi, and Tianyuan Su. Collective wisdom: Policy averaging with an application to the newsvendor problem, 2025. URL https://arxiv.org/abs/2503.17638

    2025 arXiv

  10. [18]

    Distributionally robust optimization under moment uncertainty with application to data-driven problems

    Erick Delage and Yinyu Ye. Distributionally robust optimization under moment uncertainty with application to data-driven problems. Operations research, 58 0 (3): 0 595--612, 2010

    2010

  11. [19]

    Predictive stochastic programming

    Yunxiao Deng and Suvrajeet Sen. Predictive stochastic programming. Computational Management Science, 19 0 (1): 0 65--98, 2022

    2022

  12. [20]

    Task-based end-to-end model learning in stochastic optimization

    Priya Donti, Brandon Amos, and J Zico Kolter. Task-based end-to-end model learning in stochastic optimization. Advances in neural information processing systems, 30, 2017

    2017

  13. [21]

    Multiple model-based reinforcement learning

    Kenji Doya, Kazuyuki Samejima, Ken-ichi Katagiri, and Mitsuo Kawato. Multiple model-based reinforcement learning. Neural computation, 14 0 (6): 0 1347--1369, 2002

    2002

  14. [22]

    Ensembles for continuous actions in reinforcement learning

    Siegmund Duell and Steffen Udluft. Ensembles for continuous actions in reinforcement learning. In ESANN, 2013

    2013

  15. [23]

    predict, then optimize

    Adam N Elmachtoub and Paul Grigas. Smart “predict, then optimize”. Management Science, 68 0 (1): 0 9--26, 2022

    2022

  16. [24]

    Divide-and-conquer reinforcement learning

    Dibya Ghosh, Avi Singh, Aravind Rajeswaran, Vikash Kumar, and Sergey Levine. Divide-and-conquer reinforcement learning. arXiv preprint arXiv:1711.09874, 2017

    2017 arXiv

  17. [25]

    Contextual policy transfer in reinforcement learning domains via deep mixtures-of-experts

    Michael Gimelfarb, Scott Sanner, and Chi-Guhn Lee. Contextual policy transfer in reinforcement learning domains via deep mixtures-of-experts. In Uncertainty in Artificial Intelligence, pp.\ 1787--1797. PMLR, 2021

    2021

  18. [26]

    Reinforcement learning with competitive ensembles of information-constrained primitives

    Anirudh Goyal, Shagun Sodhani, Jonathan Binas, Xue Bin Peng, Sergey Levine, and Yoshua Bengio. Reinforcement learning with competitive ensembles of information-constrained primitives. arXiv preprint arXiv:1906.10667, 2019

    1906 arXiv

  19. [27]

    A data-driven newsvendor problem: From data to decision

    Jakob Huber, Sebastian M \"u ller, Moritz Fleischmann, and Heiner Stuckenschmidt. A data-driven newsvendor problem: From data to decision. European Journal of Operational Research, 278 0 (3): 0 904--915, 2019

    2019

  20. [28]

    Stochastic optimization forests

    Nathan Kallus and Xiaojie Mao. Stochastic optimization forests. Management Science, 69 0 (4): 0 1975--1994, 2023

    1975

  21. [29]

    Residuals-based distributionally robust optimization with covariate information

    Rohit Kannan, G \"u zin Bayraksan, and James R Luedtke. Residuals-based distributionally robust optimization with covariate information. Mathematical Programming, 207 0 (1): 0 369--425, 2024

    2024

  22. [30]

    The single-period (news-vendor) problem: literature review and suggestions for future research

    Moutaz Khouja. The single-period (news-vendor) problem: literature review and suggestions for future research. omega, 27 0 (5): 0 537--553, 1999

    1999

  23. [31]

    The sample average approximation method for stochastic discrete optimization

    Anton J Kleywegt, Alexander Shapiro, and Tito Homem-de Mello. The sample average approximation method for stochastic discrete optimization. SIAM Journal on optimization, 12 0 (2): 0 479--502, 2002

    2002

  24. [32]

    Robust stochastic approximation approach to stochastic programming

    Arkadi Nemirovski, Anatoli Juditsky, Guanghui Lan, and Alexander Shapiro. Robust stochastic approximation approach to stochastic programming. SIAM Journal on optimization, 19 0 (4): 0 1574--1609, 2009

    2009

  25. [33]

    Prescriptive analytics for flexible capacity management

    Pascal M Notz and Richard Pibernik. Prescriptive analytics for flexible capacity management. Management Science, 68 0 (3): 0 1756--1775, 2022

    2022

  26. [34]

    Applying deep learning to the newsvendor problem

    Afshin Oroojlooyjadid, Lawrence V Snyder, and Martin Tak \'a c . Applying deep learning to the newsvendor problem. Iise Transactions, 52 0 (4): 0 444--463, 2020

    2020

  27. [35]

    Integrated conditional estimation-optimization

    Meng Qi, Paul Grigas, and Zuo-Jun Max Shen. Integrated conditional estimation-optimization. arXiv preprint arXiv:2110.12351, 2021

    2021 arXiv

  28. [36]

    A stochastic approximation method

    Herbert Robbins and Sutton Monro. A stochastic approximation method. The annals of mathematical statistics, pp.\ 400--407, 1951

    1951

  29. [37]

    A survey of contextual optimization methods for decision-making under uncertainty

    Utsav Sadana, Abhilash Chenreddy, Erick Delage, Alexandre Forel, Emma Frejinger, and Thibaut Vidal. A survey of contextual optimization methods for decision-making under uncertainty. European Journal of Operational Research, 320: 0 271--289, 2025. doi:10.1016/j.ejor.2024.03.020

    2025 doi

  30. [38]

    Inter-module credit assignment in modular reinforcement learning

    Kazuyuki Samejima, Kenji Doya, and Mitsuo Kawato. Inter-module credit assignment in modular reinforcement learning. Neural Networks, 16 0 (7): 0 985--994, 2003

    2003

  31. [39]

    Monte carlo sampling methods

    Alexander Shapiro. Monte carlo sampling methods. Handbooks in operations research and management science, 10: 0 353--425, 2003

    2003

  32. [40]

    On complexity of stochastic programming problems

    Alexander Shapiro and Arkadi Nemirovski. On complexity of stochastic programming problems. In Continuous optimization: Current trends and modern applications, pp.\ 111--146. Springer, 2005

    2005

  33. [41]

    Lectures on stochastic programming: modeling and theory

    Alexander Shapiro, Darinka Dentcheva, and Andrzej Ruszczynski. Lectures on stochastic programming: modeling and theory. SIAM, 2021

    2021

  34. [42]

    Facility location under uncertainty: a review

    Lawrence V Snyder. Facility location under uncertainty: a review. IIE transactions, 38 0 (7): 0 547--564, 2006

    2006

  35. [43]

    Fundamentals of supply chain theory

    Lawrence V Snyder and Zuo-Jun Max Shen. Fundamentals of supply chain theory. John Wiley & Sons, 2019

    2019

  36. [44]

    Switching between different state representations in reinforcement learning

    Harm Van Seijen, Bram Bakker, Leon Kester, et al. Switching between different state representations in reinforcement learning. In Proceedings of the 26th IASTED International Conference on Artificial Intelligence and Applications, pp.\ 226--231, 2008

    2008

  37. [45]

    Ensemble algorithms in reinforcement learning

    Marco A Wiering and Hado Van Hasselt. Ensemble algorithms in reinforcement learning. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 38 0 (4): 0 930--936, 2008

    2008

  38. [46]

    Assessing the performance of deep learning algorithms for newsvendor problem

    Yanfei Zhang and Junbin Gao. Assessing the performance of deep learning algorithms for newsvendor problem. In International conference on neural information processing, pp.\ 912--921. Springer, 2017

    2017

  39. [47]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.