{"id":"18bedc98-4a88-4168-a56c-be07270ef3b9","arxiv_id":"2508.06965","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Proposes decision-dependent distributionally robust optimization with finite-sample guarantees and a tractable reformulation, demonstrated on dynamic pricing.","lead":"This paper introduces a decision-dependent distributionally robust optimization method, where the ambiguity set adapts to the decision using interpolation and Wasserstein distances. It provides finite-sample guarantees and demonstrates the approach on dynamic pricing with nonstationary demand.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Containment guarantee silently depends on offline decision points covering the decision domain; with observational data this can fail.","rationale":"The reader's weakest assumption (smoothness of x↦P_x) is necessary but not sufficient. Even a perfectly smooth family of distributions cannot be reconstructed at a point x where offline data supplies no nearby decision points. Interpolation error is controlled only if the offline design has bounded fill distance over the domain. The abstract emphasizes 'finite offline dataset' but does not specify how those decision points are obtained. If they are observational (e.g., historical prices from a business cycle), coverage can be poor, making the containment guarantee either false or trivially weak (radius grows to infinity). The full text might state a density assumption, but based solely on the abstract this is the most load-bearing unstated condition. My concern does not overturn the reader's UNVERDICTED verdict; it reinforces that the verdict should remain UNVERDICTED pending verification of the proof's coverage assumptions and of whether the numerical experiments actually stress-test extrapolation.","tokens_in":561,"tokens_out":3029,"duration_ms":33169,"concrete_test":"Check the proof of the containment theorem (if available) for a condition of the form: decision points {x_i} are ε-dense in the domain X, and the Wasserstein radius ρ_n(x) grows with the local fill distance and the modulus of continuity of x↦P_x. Then run a numerical stress test with a 1D pricing problem: sample historical prices only in [0,0.2]∪[0.8,1], with true demand distribution varying linearly in between; compute the constructed ambiguity set at x=0.5 and see whether it contains the true distribution with the stated probability. If not, the guarantee lacks the coverage assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—finite-sample, high-probability containment of the true decision-dependent distribution in the interpolated Wasserstein ambiguity set—rests on the interpolation step. From finite pairs (x_i, ξ_i), the authors form μ̂_x for every x. A uniform containment bound requires the interpolation error d_W(μ̂_x, P_x) to be controlled with high probability for all x. The abstract does not state the needed control. Even granting the smoothness (Hölder) assumption on x↦P_x that the reader flags, a separate condition is required: the offline decision points must be sufficiently dense in the decision domain (bounded fill distance). If the dataset is observational, historical decisions may cover only a narrow region or leave gaps; in any uncovered region, μ̂_x is an extrapolation, and no finite Wasserstein radius can guarantee containment unless the radius is allowed to grow without bound. The paper's guarantee is therefore vacuous or false without an explicit coverage/design assumption, which is absent from the abstract.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a decision-dependent distributionally robust optimization (DD-DRO) framework for settings where the uncertain parameter distribution depends on the decision variable and is observed only through a finite offline dataset. It constructs decision-dependent nominal distributions by multivariate interpolation and builds Wasserstein ambiguity sets around them. The abstract claims finite-sample high-probability containment of the true decision-dependent distribution, a non-asymptotic out-of-sample guarantee, an optimality gap bound, and a tractable reformulation, with numerical validation on dynamic pricing.","tokens_in":799,"tokens_out":3179,"duration_ms":32488,"significance":"The DDU problem is important and the interpolation-based approach is a plausible way to turn offline data into decision-dependent ambiguity sets. If the stated guarantees hold, the contribution is meaningful for data-driven decision-making under distribution shift caused by decisions. However, because the manuscript was provided to this referee as an abstract only, the derivation, assumptions, and proofs cannot be independently checked. No code, theorem statements, or appendices are available to verify the claimed uniform containments and tractability.","major_comments":[{"comment":"The finite-sample containment guarantee is stated without any coverage assumption on the offline decision points. For every decision x, μ̂_x is built by interpolating finite pairs (x_i, ξ_i). A uniform high-probability bound on d_W(μ̂_x, P_x) requires control of the fill distance of {x_i} in the decision domain (or a suitable subset). With observational data, historical decisions may leave regions uncovered; in those regions the guarantee is vacuous or false unless the Wasserstein radius is allowed to grow. The abstract must state the design/coverage assumption or qualify the claim. As written, this is a load-bearing omission.","section":"Abstract"},{"comment":"The non-asymptotic out-of-sample performance guarantee and optimality gap bound are asserted but none of the underlying regularity conditions (e.g., Hölder smoothness of x↦P_x, bounded support, light tails, sample-size and radius scaling) are reported. It is impossible to assess whether the bounds are non-trivial or merely hold with radii that absorb all interpolation error. The authors should state the precise dependence of the radius on sample size, interpolation error, and confidence.","section":"Abstract"},{"comment":"Tractability of the DD-DRO reformulation is claimed but not specified. Whether the reformulation is convex, finite-dimensional, or solvable by standard methods matters for the practical claim; in DDU problems even the nominal distribution's decision dependence can break convexity. The full text must provide the reformulation and a statement of assumptions under which it is tractable.","section":"Abstract"}],"minor_comments":[{"comment":"'Decision-dependent nominal distributions (thereby decision-dependent ambiguity sets)' is confusing; clarify the construction order.","section":"Abstract"},{"comment":"'Guaranteed expected revenue' in the last sentence is too strong if the guarantee is probabilistic; suggest 'high-probability expected revenue guarantee'.","section":"Abstract"},{"comment":"No references to prior DDU or data-driven DRDO work appear in the abstract; the introduction should supply the necessary context.","section":"Abstract"}],"recommendation":"uncertain","confidential_remarks":"I was provided only the abstract. The full text is required before I can issue a definitive recommendation. The coverage concern from the stress-test note is genuine and should be addressed in the revision: the abstract's containment guarantee needs an explicit coverage/fill-distance assumption or a precise statement of how the radius grows outside the support of the historical decisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"We only have the abstract, so this is a first read, not a verdict on the math. The core idea is a good one: instead of building one nominal distribution from data, interpolate the nominal distribution as a function of the decision variable, then wrap it in a Wasserstein ball. As far as I can tell from the abstract, that combination of decision-dependent uncertainty with interpolation-based Wasserstein DRO is new, and the dynamic pricing application gives it a natural testbed. The paper does the right things: it states finite-sample guarantees, an out-of-sample performance bound, an optimality gap, and tractability, and it backs the formulation with numerical experiments.\n\nThe soft spot is right where the stress-test note lands. The claim that the ambiguity sets contain the true decision-dependent distribution with high probability requires a uniform bound on the interpolation error d_W(μ̂_x, P_x) over all x. A uniform bound requires two things: smoothness of x ↦ P_x (Hölder or similar), and — critically — that the offline decision points are dense enough in the decision domain (a bounded fill distance). The abstract mentions neither. If the data are observational, historical decisions can easily leave large gaps or only cover a narrow regime; then μ̂_x is an extrapolation in the uncovered regions, and no finite Wasserstein radius can guarantee containment. That isn't a fatal flaw if the full paper states and uses a coverage/design assumption, but the abstract as written sells the guarantee without it. That's a real omission, and it should be surfaced in any review.\n\nI'd also want to see how they handle the interpolation method and the constant in the radius. But those are details. The main question is whether the high-probability containment is truly uniform in x or only pointwise; the abstract reads as uniform, which is the strong claim.\n\nWho is this for? People working on DRO with endogenous uncertainty, and on data-driven dynamic pricing under nonstationary demand. If the proofs deliver, it's a useful tool. I can't cite it yet without seeing the coverage assumption and the uniform bound, but I would send it to a competent referee. The authors seem to know the literature and are not overclaiming beyond what the abstract states — the gap is in what they don't state.\n\nBottom line: worth a serious referee. The referee should press on the coverage condition and on whether the finite-sample guarantee is uniform over the decision domain.","headline":"A plausible and timely combination of DDU and Wasserstein DRO, but the finite-sample containment guarantee silently needs an explicit coverage/design condition on the offline decision points; the abstract omits it.","tokens_in":1132,"tokens_out":1537,"would_cite":false,"duration_ms":18542,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["90C15","90C47"],"pacs":[],"model":"deepseek-v4-flash","headline":"A finite offline dataset suffices to build decision-dependent Wasserstein ambiguity sets with high-probability containment, and the resulting robust problem is tractable with non-asymptotic guarantees.","keywords":["decision-dependent uncertainty","distributionally robust optimization","Wasserstein metric","multivariate interpolation","finite-sample guarantee","dynamic pricing","nonstationary demand","non-asymptotic performance bound"],"falsifier":"Simulate a decision-dependent uncertainty model in which the distribution of the uncertain parameter is a discontinuous or highly oscillatory function of the decision, estimate the nominal family from a finite offline sample using the paper's interpolation scheme, and check the empirical frequency with which the true distribution leaves the claimed Wasserstein ball; if that frequency systematically exceeds the stated confidence level, the finite-sample containment claim is false in that regime.","tokens_in":540,"feed_emoji":"📈","tokens_out":8596,"duration_ms":73656,"temperature":0.7,"pith_summary":"Standard distributionally robust optimization assumes one fixed nominal distribution, but in many decision problems the uncertainty itself shifts with the chosen decision—demand falls when price rises, for example. This paper tackles that decision-dependent uncertainty by building a family of nominal distributions from a finite offline dataset, using multivariate interpolation to make the nominal distribution a function of the decision and the Wasserstein metric to define a ball of plausible distributions around each decision point. The central claim is that these decision-dependent ambiguity sets contain the true distribution with finite-sample high probability, and that the resulting min-max problem is tractable and comes with non-asymptotic out-of-sample and optimality-gap guarantees. The authors test the framework on a dynamic pricing problem with nonstationary demand, obtaining pricing strategies with guaranteed expected revenue.","feed_headline":"Pricing under decision-dependent demand gets a data-driven guarantee","feed_subtitle":"Wasserstein-based ambiguity sets from offline data give provable performance guarantees.","key_machinery":"The central mechanism is multivariate interpolation over the decision space combined with the Wasserstein metric. The offline data are used to estimate the conditional distribution of the uncertain parameter at the sampled decision points; an interpolation scheme extends these estimates to every decision $x$, giving a nominal distribution $\\hat{P}(x)$. A Wasserstein ball of radius $\\varepsilon$ around $\\hat{P}(x)$ defines the ambiguity set. The radius is the single tuning parameter that absorbs both sampling error and interpolation error, and the paper's finite-sample bound shows how to choose it so that the true distribution is trapped with high probability. The tractable reformulation foll","core_discovery":"The paper's central discovery is that decision-dependent ambiguity sets can be constructed directly from offline data rather than assumed known. By interpolating the empirical distributions observed at different decision points and placing a Wasserstein ball around each interpolated nominal distribution, the authors obtain a set of distributions $\\mathcal{P}(x)$ for each decision $x$ such that, with high probability over the offline sample, the true conditional distribution $P_x$ lies in $\\mathcal{P}(x)$. They further show the DD-DRO problem $\\min_{x \\in X} \\sup_{P \\in \\mathcal{P}(x)} \\mathbb{E}_P[c(x,\\xi)]$ has a tractable reformulation, and they derive finite-sample out-of-sample performan","pith_inferences":["A natural next question, not addressed in the abstract, is how the Wasserstein radius must grow with the dimension of the decision space and the sparsity of the offline sample; interpolation error typically worsens in higher dimensions, so the finite-sample bounds may become conservative in large problems.","The same construction could be transferred to other decision-dependent settings—personalized pricing with customer features, inventory control with price-dependent demand, or network routing with congestion—wherever offline data can be used to estimate how a distribution moves with the decision.","A direct test would be to compare this decision-dependent ambiguity set against a pooling or decision-independent DRO benchmark and measure, in simulation, which one attains closer to the true optimal objective when the distribution actually shifts with the decision."],"forward_implications":["Finite offline data suffice to certify, with high probability, that the true decision-dependent distribution sits inside the ambiguity set, so the robust decision is protected against distributional misspecification that other data-driven DRO methods ignore.","The tractable reformulation means the robust pricing (or other) policy can be computed in practice using standard optimization techniques rather than requiring online distribution estimation.","The non-asymptotic out-of-sample guarantee tells a decision maker how many offline samples are needed to achieve a target level of worst-case performance.","The optimality gap bound quantifies the cost of robustness: it bounds how much the worst-case objective of the robust solution exceeds the true optimal value.","For dynamic pricing with nonstationary demand, the framework yields pricing strategies that carry a guaranteed expected revenue across all distributions in the ambiguity set."],"supporting_citations":[],"fun_headline_variants":["Data-driven ambiguity sets for decision-dependent pricing","Wasserstein-based sets guarantee robust dynamic pricing","Offline data yields provable guarantees for shifting demand","Decision-dependent pricing gets finite-sample guarantees"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The true decision-dependent distribution must vary smoothly enough with the decision that multivariate interpolation from a finite offline dataset can approximate the unobserved intermediate distributions well; if the distribution changes sharply or irregularly with the decision, the nominal family is inaccurate and the coverage and performance guarantees do not hold.","fun_headline_variants_meta":{"raw":{"variants":["Data-driven ambiguity sets for decision-dependent pricing","Wasserstein-based sets guarantee robust dynamic pricing","Offline data yields provable guarantees for shifting demand","Decision-dependent pricing gets finite-sample guarantees"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000263,"raw_usage":{"total_tokens":1406,"prompt_tokens":686,"completion_tokens":720,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":430,"completion_tokens_details":{"reasoning_tokens":663}},"tokens_in":430,"tokens_out":720,"duration_ms":6985,"temperature":1.0,"reasoning_tokens":663,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T22:23:29.863522+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate a decision-dependent uncertainty model in which the distribution of the uncertain parameter is a discontinuous or highly oscillatory function of the decision, estimate the nominal family from a finite offline sample using the paper's interpolation scheme, and check the empirical frequency with which the true distribution leaves the claimed Wasserstein ball; if that frequency systematically exceeds the stated confidence level, the finite-sample containment claim is false in that regime.","supporting_citations":[],"review_version":1}