Pith. sign in

REVIEW 8 cited by

Estimate-Then-Optimize versus Integrated-Estimation-Optimization versus Sample Average Approximation: A Stochastic Dominance Perspective

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2304.06833 v4 pith:T3L5SCNW submitted 2023-04-13 stat.ML cs.LGstat.ME

classification stat.MLcs.LGstat.ME
keywords modelwhenoptimizationclassstochasticregretapproximationaverage
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

In data-driven stochastic optimization, model parameters of the underlying distribution need to be estimated from data in addition to the optimization task. Recent literature considers integrating the estimation and optimization processes by selecting model parameters that lead to the best empirical objective performance. This integrated approach, which we call integrated-estimation-optimization (IEO), can be readily shown to outperform simple estimate-then-optimize (ETO) when the model is misspecified. In this paper, we show that a reverse behavior appears when the model class is well-specified and there is sufficient data. Specifically, for a general class of nonlinear stochastic optimization problems, we show that simple ETO outperforms IEO asymptotically when the model class covers the ground truth, in the strong sense of stochastic dominance of the regret. Namely, the entire distribution of the regret, not only its mean or other moments, is always better for ETO compared to IEO. Our results also apply to constrained, contextual optimization problems where the decision depends on observed features. Whenever applicable, we also demonstrate how standard sample average approximation (SAA) performs the worst when the model class is well-specified in terms of regret, and best when it is misspecified. Finally, we provide experimental results to support our theoretical comparisons and illustrate when our insights hold in finite-sample regimes and under various degrees of misspecification.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

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

  1. Integrated Learning and Robust Optimization

    math.OC 2026-08 conditional novelty 7.0 of 10

    A new framework, ILRO, trains cost predictors through a robust linear program, with a convex surrogate and consistency and convergence guarantees under stated conditions.

  2. End-to-End Fairness Optimization with Fair Decision-Focused Learning

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Jointly optimizing prediction accuracy, prediction disparity, and decision regret during training yields fairer prediction-informed resource allocations, with closed-form decision Jacobians for α-fair allocation problems.

  3. Weak-to-Strong Learning in Decision Making

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Weak-to-strong training with pseudo-distributions can improve downstream decision risk over strong-only training when labels are scarce, unlabeled data are abundant, and weak/strong feature overlap is small.

  4. Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents

    cs.LG 2025-02 conditional novelty 6.0 of 10

    Prediction sets paired with max-min decisions are a sufficient statistic for value-at-risk optimizing agents, and the RAC algorithm builds such sets with distribution-free coverage.

  5. Pessimistic bilevel optimization approach for decision-focused learning

    math.OC 2025-01 conditional novelty 6.0 of 10

    A branch-and-cut method minimizes the pessimistic IEO regret loss directly for 0-1 combinatorial decision-focused learning, avoiding the need for a convex hull.

  6. DFF: Decision-Focused Fine-tuning for Smarter Predict-then-Optimize with Limited Data

    cs.LG 2025-01 conditional novelty 6.0 of 10

    Decision-Focused Fine-tuning corrects any backbone predictor's outputs within a bounded trust region using a residual scaling layer, improving decision regret on predict-then-optimize tasks.

  7. Toward Decision-Oriented Prognostics: An Integrated Estimate-Optimize Framework for Predictive Maintenance

    cs.AI 2025-06 conditional novelty 5.0 of 10

    Training predictive maintenance models on downstream maintenance cost instead of prediction accuracy alone lowers average maintenance regret in turbofan experiments, by up to about 22% in the long-term setting.

  8. Towards Foundation-model-based Multiagent System to Accelerate AI for Social Impact

    cs.AI 2024-12 unverdicted novelty 5.0 of 10

    The paper outlines a research vision for using LLM-based meta-agents to automate problem formulation, solution design, and evaluation in AI for social impact.

Pith tools