| paper | d4-journals-mss-2020-0029 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | no |
Theorem 1 and Theorem 2 compare solution concepts, and Proposition 1 characterizes incentive efficiency. None asserts that computing a solution is tractable or hard, or provides a formulation used by the paper to compute one, so bit (a) fails.
fails bit a — no named computational result to mirror
No result is covered by a computational mirror: Theorem 1 and Theorem 2 establish solution-concept equivalences, while Proposition 1 characterizes incentive efficiency.
I don’t find a qualifying anchor in this paper. It contains named mathematical results, but none asserts a complexity classification such as polynomial-time solvability, NP-hardness, or fixed-parameter tractability.
The closest hooks are Theorem 1, proved in the paper: for Bayesian cooperative games with orthogonal coalitions and sidepayments, the sets of interim utility allocations from Myerson’s M-solution and Salamanca’s S-solution are equal. Theorem 2, also established in the paper, says that in two-player Bayesian games with sidepayments and verifiable types, the MS-solution and the cooperative-competitive value are ex-ante utility equivalent. Both are results about solution concepts, not about the computational complexity of finding or evaluating them. Proposition 1 characterizes incentive-efficient mechanisms through virtual utilities, but it likewise gives no complexity result.
The paper’s probability distribution \(p\) is over the possible type profiles of a finite set of players. It is not mass over a population of interchangeable agents. Turning it into a population distribution would require changing the model: coalitions and the Shapley value depend on the identities and subsets of players, and the paper supplies no continuum-player version of those objects.
A large cooperative organization with many otherwise identical members might motivate a new continuous-population model. But the paper’s named results do not establish computational claims about that model, so I can’t state a continuous problem per qualifying anchor without inventing a connection. Under the requested standard, the honest positive case is therefore empty.
The proponent’s conclusion is right, but the three named results should be ruled out one by one. The decisive issue is that none states a computational problem or a complexity result. Theorem 1 equates two sets of interim utility allocations; Theorem 2 establishes ex-ante utility equivalence; Proposition 1 characterizes incentive efficiency through virtual utilities. A question about whether a continuous version preserves any of these statements would be a question about solution concepts or axioms, not the computational landscape ChoCo studies. “Compute an MS-value” could be an algorithmic problem, but this paper neither formulates that task nor analyzes its complexity. Adding it would create a new research question rather than mirror one of the paper’s results.
Theorem 1 is the strongest candidate for a population extension: one might ask whether M- and S-values still coincide in a continuum-player cooperative game. But that asks whether an equivalence between solution concepts survives, not how hard it is to find a solution. And a continuum-player version would need its own definitions of coalitions, value allocation, and the limiting role of the Shapley value. A useful theory might result, but the theorem supplies no computational target for it.
Theorem 2 is harder to turn into a population mirror. It is specifically about two-player games. Repeating many copies of a two-player game gives a population of bargaining instances, while replacing the two players with a large population creates a different game in which the theorem’s two-player comparison no longer applies. Either might inspire research, but neither makes this theorem a computational result about a continuous society.
Proposition 1 has an optimization flavor, but its distribution \(p\) describes uncertainty about the types of a fixed set of players. It is not a distribution of interchangeable agents. Its maximization over decisions \(d\) could motivate an algorithmic study if \(D\) had a compact, difficult representation; the paper instead assumes finite decision sets and gives no complexity analysis. That possible pricing question would concern the decision space, not population continuization.
A large cooperative organization with many players sharing the same relevant characteristics could motivate a continuum cooperative-game model. That is the best counterexample to a blanket claim that no related project could be worthwhile. But the model would have to specify how coalition formation, player multiplicities, and value allocation scale, and the paper gives no computational result for it to mirror. I would therefore defend the negative verdict for this paper as a ChoCo target, while not claiming that continuous cooperative-game computation is inherently uninteresting.
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.