| paper | d4-conf-sagt-2011-0010 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | no |
The paper’s numbered results are an axiomatic uniqueness characterization, an identity with McQuillin’s value, and an algebraic basis lemma; none asserts a computational result under the stated test. Since bit (a) fails, no continuous mirror of a result in this paper qualifies.
fails bit a — no named computational result to mirror
The paper’s uniqueness characterization, equivalence with McQuillin’s value, and basis result are all axiomatic or algebraic; none supplies a computational result to mirror.
The strongest honest case for a continuous mirror is limited: this paper has no named computational result to anchor one to. Theorem 5.1, proved here, uniquely characterizes a value satisfying Efficiency, Symmetry, Linearity, and the steady-marginality Null-player Axiom. Theorem 5.4, also proved here, shows that this value equals McQuillin’s value. Lemma 4.4 supplies a basis for the proof. These are axiomatic and algebraic results; none asserts tractability, hardness, or an algorithm. Theorem 5.4 is an identity between two values on the paper’s finite-agent model, not a discrete-to-continuous identity.
The closest plausible population regime would be a large consortium of interchangeable suppliers: finitely many complete contract, cost, or technology types, each held by a fraction \(\mu_t\) of the population. A possible new question would represent each coalition by its type-mass vector and ask for the McQuillin-style payoff rate for each type. But an infinitesimal agent’s transfer may have no effect in an ordinary nonatomic model, so defining that value would require new machinery, such as directional derivatives. That would be a continuous computational question inspired by the paper, not a mirror of a named computational result in it.
I therefore can’t responsibly name a lead anchor or give the requested per-anchor problem. The weakness is decisive for this task: the paper studies axiomatic characterization, and its results do not establish a computational problem whose population continuization we can assess.
The strongest case against a ChoCo mirror is that the paper has no computational result to continuize. Theorem 5.1 characterizes a value by axioms; Theorem 5.4 identifies it with McQuillin’s value; Lemma 4.4 proves an algebraic basis result. None studies an algorithm, a complexity class, or a computational task. That absence is decisive for this programme’s purpose.
The supplier consortium is a plausible high multiplicity setting, and the proponent’s proposed type payoff rate could be made coherent. One could represent coalitions by type masses and replace the effect of moving a single agent with a directional derivative. But that produces a continuous value theory: it asks how to define or characterize an allocation, not how the complexity of a computational problem changes when the population becomes continuous. A computational version would need an additional task and a finite representation of the externality function. The paper supplies neither. With the full partition function listed explicitly, the McQuillin value can be computed by the ordinary Shapley sum in time polynomial in that table’s size; with a succinct representation, the complexity depends on a representation the paper never chooses.
Theorem 5.4 is the closest candidate for a “turns out” result, but it equates two values on the same finite-agent model. It does not show that a discrete stochastic model converges to a continuous one, or that finite and mass-based formulations coincide. A continuous analogue would require defining both steady marginality and the comparison value for infinitesimal transfers. If those definitions yield an identity, that is still an axiomatic identity, not a computational result. The basis lemma has the same problem: a continuous function-space analogue would be algebraic, and the paper gives no computational question for it to answer.
So the paper can motivate a new continuous cooperative-game project, perhaps one with computational questions added later. The honest weakness in the negative case is that no one can rule out such a project being worthwhile in its own right. But making it a computational project requires choices about tasks and input representations that are absent here. That is a new programme inspired by the paper, not a worthwhile continuous mirror of one of its results.
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.