| paper | d4-journals-jair-2013-0038 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | unclear |
The paper’s only numbered theorem characterizes the distributional form implied by its axioms; it makes no computational claim. Section 5’s regulator optimization and the proposed portfolio problem may motivate computational work, but neither is a named computational result of this paper.
fails bit a — no named computational result to mirror
No computational theorem is covered; Theorem 1 supplies a distributional form, while Sections 4–5 demonstrate prediction and regulator optimization without complexity guarantees.
The strongest positive mirror I can make is a high-multiplicity version of the regulator problem in Section 5, applied to a portfolio of recurring flight rerouting negotiations. But the paper has no numbered computational result asserting a complexity classification. Its only numbered theorem is analytic, so this cannot be a fully compliant anchored case.
The closest named result is Theorem 1, proved in this paper. It says that for an \(N\)-dimensional NDB concept satisfying the theorem’s differentiability assumption, there are constants \(a_i\), independent of the bargaining problem \((S,d)\), such that for every \(x\in S\) with \(x\succ d\),
\[
\mu_{S,d}(x)\propto\prod_{i=1}^{N}(x_i-d_i)^{a_i}.
\]
That gives a computable form for the distribution used in the regulator’s expected-welfare calculation. It does not assert that a computational problem is in P, hard, or otherwise classify its complexity. The closest computational hook is Section 5’s optimization over regulator interventions—especially Eq. (6)—and its flight-rerouting example in Figure 4; those are not numbered theorem, lemma, corollary, or proposition results.
The lead candidate mirror is High-Multiplicity NDB Portfolio Management. Imagine many recurring flight-level cockpit–ATC negotiations: for example, flights in repeated route and operating-condition cohorts. A type \(t\) must specify everything the problem uses: its feasible joint-utility set, disagreement point, bargaining exponents, regulator’s welfare function, available interventions, and their costs. Mass \(\mu_t\) is the fraction of negotiations of that type. The regime is plausible when the number of flights is large and many share a relatively small set of standardized environments; it is much less plausible when each flight’s weather, traffic, or utility data makes it a distinct type.
Precisely, an instance gives a finite support of types with rational masses summing to one. For each type \(t\), it gives a finite menu \(R_t\) of regulator interventions. Intervention \(r\) specifies a finite rational outcome set \(S_{tr}\), disagreement point \(d_{tr}\), rational welfare values \(U_t(x)\), and rational cost \(c_{tr}\). Take the NDB exponents to be fixed positive integers and use the paper’s flight-model convention that outcomes not strictly better than disagreement have probability zero. For \(x\succ d_{tr}\), define
\[
p_{tr}(x)=
\frac{\prod_i(x_i-d_{tr,i})^{a_i}}
{\sum_{y\in S_{tr}:y\succ d_{tr}}\prod_i(y_i-d_{tr,i})^{a_i}}.
\]
Given a total intervention budget \(B\), the question is to choose masses \(z_{tr}\ge 0\) assigned to each type–intervention pair, maximizing
\[
\sum_{t,r}z_{tr}\sum_{x\in S_{tr}:x\succ d_{tr}}p_{tr}(x)U_t(x)
\]
subject to \(\sum_{r\in R_t}z_{tr}=\mu_t\) for every type and \(\sum_{t,r}c_{tr}z_{tr}\le B\). A solution is an optimal assignment of intervention mass \(z\). This is the continuous relaxation of assigning interventions to a finite portfolio of flight negotiations, where the discrete version restricts assigned masses to multiples of \(1/n\).
The authors would likely recognize the per-negotiation objective: choosing an intervention to maximize expected welfare is exactly the regulator’s task in Section 5. The portfolio and budget add a population-level allocation question around that task. With explicit finite menus and outcome sets, the expected-welfare coefficients can be computed from Theorem 1’s formula, and the remaining problem is a linear program. I would expect this mirror to be tractable in that explicit-input regime. It also raises questions about compactly represented intervention sets, shared interventions across types, and network effects between negotiations.
This covers the paper’s NDB management application, not its axiomatic derivation as a population problem. The weak point is substantial: the paper studies intervention for a bargaining instance, while the portfolio and budget are an added scale-up. Whether the authors accept that as their problem’s continuous form depends on whether they recognize managing repeated negotiations as part of the same regulator problem. Theorem 1 supports the formulation, but it cannot serve as the required computational anchor.
The proponent’s named anchor, Theorem 1, is an axiomatic characterization of the distribution over outcomes of a single bargaining problem:
\[
\mu_{S,d}(x)\propto\prod_i(x_i-d_i)^{a_i}.
\]
It gives no complexity classification, computational problem, or algorithmic bound. Section 5’s regulator equations describe an optimization task, but the paper does not analyze its computational complexity. So the proposed portfolio problem is not a continuous mirror of a computational result in this paper.
The strongest version of that proposal is still coherent: treat recurring flight negotiations as a population of types, with each type specifying its feasible outcomes, disagreement point, utilities, and available interventions. If the regulator chooses one common policy, the population simply aggregates the expected welfare calculations for individual instances. If the regulator has a shared budget and can assign different interventions to different fractions of each type, the budget couples those choices. That is a sensible control problem over a population of negotiation instances.
Its weakness as a *mirror of this paper* is that the portfolio and its shared budget supply the central computational structure. The paper studies a regulator changing one bargaining game; it does not study allocating scarce interventions across games. The proposed finite-menu model makes the added structure especially visible: it turns the task into a mass-allocation problem using expected-welfare values from the paper. That may be useful work, but Theorem 1 does not establish its algorithmic question.
I would not reject this candidate on the grounds that flights are individually varied, that the population is too small, or that the resulting optimization is tractable. A type can include all problem-relevant flight conditions, and recurring cohorts are plausible. Nor can I confidently say the authors would not recognize portfolio management as an extension of their regulator problem. The negative case is strong against the proponent’s *anchor*: no named computational result in the paper supports it. It is weak against the broader claim that no worthwhile continuous mirror exists; the recurring-negotiation control problem could be worthwhile if the authors accept it as their problem.
The adversarial triple: the proponent anchors on up to three named results; the opponent sees that case and must defeat every anchor; the judge decides which case convinced it. These are the pipeline’s own outputs, generated by tools/triple_run.py — no human edited them. The paper’s own text is not reproduced here beyond the quoted statement above.