| paper | d4-conf-atal-2017-0053 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | medium |
| authors would recognise it | no |
Theorems 2 and 3 provide explicit equilibrium characterizations, so the lack of a complexity classification alone does not rule them out as computational anchors. But a continuum of strategic contributors with graphon referral flows would replace the finite-player Nash and subgame-perfect equilibria with a nonatomic network-game solution concept; the paper does not specify that limit or a computational task for it.
fails bit b-changes — the continuum limit changes the question
The candidate anchors are Theorem 2 for REPP-R and Theorem 3 for REPP-S; neither yields a continuous computational mirror. No other result was anchored.
I can’t give a qualifying positive case under the anchor rule: this paper has no named computational complexity result. Theorem 2 and Theorem 3 are the nearest candidates, but they characterize equilibria; they do not show that a problem is in P, NP-hard, or otherwise classify its computational complexity.
The strongest plausible setting for a population mirror would be a large civic project with many residents grouped by identical valuation and role in a repeated referral network. A type would include valuation, arrival cohort, and referral opportunities; its mass would be the fraction of residents of that type. The paper’s Theorem 2 proves that, under its referral-bonus conditions and the stated bound on \(\sigma\), REPP-R has Nash equilibria in which agents refer their neighbors and aggregate contributions meet the provision point. Theorem 3 gives the sequential counterpart for REPP-S: under its cost-function conditions and bound on \(\sigma\), there are subgame-perfect equilibria that fund the project, with agents referring their neighbors and contributing upon arrival. Both are proved in this paper.
One could formulate a continuous equilibrium-computation problem for such type distributions and referral flows. But the paper does not study that problem computationally, and its results rely on the referral network’s topology and diameter. Capturing those features in a type-level model would require substantial additional modeling. That could motivate a follow-up, but it is not an honest mirror of a named computational result here. The gap is not that crowdfunding lacks a plausible high-multiplicity scenario; it is that the paper’s results are outside the programme’s required computational anchors.
The decisive problem is that this paper has no named computational result to mirror. Theorem 2 characterizes Nash equilibria for REPP-R; Theorem 3 characterizes subgame-perfect equilibria for REPP-S. Neither studies an algorithm or a complexity question. The proponent’s proposed “continuous equilibrium-computation problem” would add a new computational task to the paper, rather than continuize one of its results.
A large civic project with many residents can be a sensible high-multiplicity setting. That answers the population-scale objection, but it does not supply the missing computational anchor. Nor is a distribution over residents’ values and arrival times enough to capture these theorems: their conditions and incentives depend on the referral network, including its connectivity, referral paths, and—in REPP-S—the history that determines securities.
The best repaired mirror would give residents a continuum of values and arrival times, plus an explicit measurable referral network or graphon and referral flows. That could be a coherent new model. But finding its equilibrium would be a new mean-field network-game question; turning it into a computational problem would additionally require an input representation and a defined algorithmic task that this paper never provides. Theorem 2’s simultaneous game and Theorem 3’s sequential game both fail for the same reason: their results concern equilibrium existence and characterization, not computational complexity.
So I would not argue that no large-population crowdfunding model could be worthwhile. The negative case is narrower and stronger: neither theorem supports a continuous *computational mirror* of this paper. A graphon extension may merit its own project, but it cannot fill a computational gap in a paper that makes no computational claim.
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.