| paper | d4-conf-ijcai-2022-0504 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | no |
Neither named anchor asserts an algorithm, a computational difficulty result, or a formulation used to compute. The paper characterizes equilibria and their accuracy, so it has no named computational result to anchor a continuous computational question.
fails bit a — no named computational result to mirror
Theorem 1 characterizes equilibria in the homogeneous game, and Theorem 6 gives accuracy results for selected graph families; neither is a computational result. The paper's other equilibrium characterizations likewise do not provide a computational anchor.
I can’t make a qualifying positive computational-continuization case from this paper: it contains no named computational-complexity result to anchor one to.
The closest result is Theorem 6, proved here: for paths, cycles, and bounded-degree trees of even branching factor and depth at least two, the accuracy in a minimum-size Nash equilibrium tends to \(1\) as the graph grows. Theorem 1, also proved here, characterizes equilibria in the equal-competence, equal-stakes setting: when \(e<p-\tfrac12\), a profile is a Nash equilibrium exactly when its number of voters is odd and at most \(n_e\). These are equilibrium and accuracy characterizations, not results placing a computational problem in P, proving hardness, or giving a computational formulation the paper solves.
A continuum network equilibrium problem could be proposed as follow-up work, but it would not be a mirror anchored in a named computational result here. It also risks losing the paper’s central mechanism: in an atomless population, an individual voter has no pivotal effect, while the paper’s equilibrium results turn on voters’ chances of breaking a tie.
The strongest objection is not that this paper studies networks, or that its conclusions might change in a continuum. It is that none of its named results poses a computational problem. Theorem 1 characterizes pure Nash equilibria in the homogeneous game; Theorem 6 characterizes equilibrium accuracy on selected graph families. Neither gives an algorithm or a complexity classification to continuize. One could ask new computational questions about these games, but that would be a follow-up to the paper’s subject, not a continuous mirror of a computational result it establishes.
Theorem 1 is the tempting anchor because it concerns a population of interchangeable voters. But its mechanism is individual pivotality: a voter matters when their ballot can break a tie. In an atomless population, one person changes voting mass by zero. If a positive mass votes, accuracy is \(1\) in the limit, so an individual can abstain and save effort without changing it; at zero voting mass, one individual still cannot change the outcome. With \(e>0\), this makes universal abstention the natural equilibrium, rather than a mass version of the paper’s odd-sized equilibria. Allowing a type to choose its voting mass instead would make types or cohorts coordinate as blocs, changing the unilateral-deviation game. A rare-turnout limit could preserve finite pivotality, but it would track a finite or Poisson number of voters as the electorate grows, not a positive mass of the society, and it would move from the paper’s pure equilibria to a different model.
Theorem 6 offers a stronger computational-looking route: on arbitrary graphs, ask for a minimum-size equilibrium or a highest-accuracy equilibrium. That is a sensible new graph problem, but its constraints come from the exact adjacency structure. A continuum over voter types does not preserve which particular nodes are neighbors. If types encode enough positional information to recover that structure, their support grows with the graph and the aggregation has lost its high-multiplicity advantage. A dense graph limit avoids that growing structure only by changing the bounded-degree local interactions on which Theorem 6 relies; a local-neighborhood distribution retains local patterns but does not by itself encode the global consistency needed to assemble an equilibrium. Repeating fixed finite motifs gives genuine multiplicity, but then equilibrium computation factors through the finite motifs; the continuum aggregates copies without resolving the graph problem the theorem is about.
So the proposed mirrors fail for different reasons: in the complete-information game, atomlessness removes pivotality; in the network game, preserving pivotality requires retaining the graph’s discrete relational structure. The paper’s conclusions about accuracy may inspire useful continuum game-theoretic work, but under ChoCo’s computational brief, those are not yet worthwhile mirrors of its named results. The negative case is not airtight—one could formulate a new computational task around arbitrary-network equilibria—but that task would need its own justification as a new problem, rather than borrowing the paper’s theorems as computational anchors.
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.