| paper | d4-conf-ecai-2016-0093 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | unclear |
The named results determine representation dimensions and construct games with large dimensions, but none gives an algorithmic guarantee or a complexity result for a computational problem. The proposed mass-based representation question is a plausible new extension, yet it introduces outcomes for fractional coalitions that the paper's games do not specify. The paper therefore fails the computational-anchor gate.
fails bit a — no named computational result to mirror
The proposed extension concerns the representation dimension studied in Theorem 1, Lemma 1, and Theorem 2; it does not provide a continuous mirror of a named computational result.
I don’t find a qualifying computational anchor in this paper. Theorem 1 (proved here) gives the exact dimension of the constructed game \(\Gamma_k\), namely \(2^{k-1}\). Lemma 1 (proved here) gives \(\dim(\Gamma_C)=|C|\), and Theorem 2 (proved here) gives lower bounds on the dimensions of large families of simple games. These are results about representation size and extremal constructions. None classifies the running time or complexity of a computational problem, so I cannot honestly use them as the required anchors.
The closest positive direction would be a new *continuum simple-game dimension* problem. Let an atomless electorate have \(\tau\) voter types, each with fixed voting rights and other relevant attributes, and let \(b_t\) be the mass of type \(t\). A coalition is described by its yes-vote masses \(x\in\prod_t[0,b_t]\). Given a rational, upward-closed winning region \(W\) in that box, ask for the smallest \(r\) such that
\[
W=\bigcap_{j=1}^{r}\left\{x:\sum_t w^j_t x_t\ge q_j\right\},
\]
with \(w^j_t\ge 0\) and \(q_j>0\); a solution is the list of thresholds, or a report that no finite representation exists. A plausible regime is a very large electorate with only a few institutional roles, so population mass is continuous while \(\tau\) stays small. This is recognizably related to the paper’s notion of dimension, though it is a proposed extension, not a mirror of one of its named computational results.
The gap is substantial: the paper’s constructions are defined on binary coalitions. In \(\Gamma_k\), for example, the outcome at the tie depends on the parity of the number of \(T\)-players in the coalition; fractional \(T\)-mass has no direct parity analogue. So Theorem 1 or Theorem 2 does not establish what happens to the dimension in this continuum problem. I would leave its complexity open; any prediction of continuum-specific hardness would be conjecture, not a conclusion supported by this paper. My strongest honest positive case is therefore that the paper suggests a plausible new continuous representation problem, while offering no qualifying computational anchor for it.
I can’t make Theorem 1, Lemma 1, or Theorem 2 qualify as computational anchors under ChoCo’s stated gate. They establish representation dimensions: the exact dimension of \(\Gamma_k\), of a code-derived game, and lower bounds for worst-case dimension. The paper does not give a complexity classification or algorithmic guarantee for a problem such as computing dimension from an input game. Its constructions are explicit, but that is not the same result as a running-time theorem.
The proposed continuous dimension problem is the strongest nearby idea, but it needs a new object. The paper’s dimension asks for agreement on the Boolean coalitions—the corners of a cube. A continuous winning region \(W\) also specifies outcomes at fractional coalitions, and the original game does not determine those outcomes. Requiring \(W\) to be an intersection of threshold regions makes the question a geometric representation problem; allowing arbitrary upward-closed \(W\) makes finite representability an additional issue. Either version could be studied, but its answers would depend on choices about the extension and input encoding.
A better attempt to retain the paper’s coding constructions runs into the type-space tradeoff. With a fixed, small set of institutional roles, coalitions are summarized by role-level masses, while the constructions use many independently distinguished player coordinates. Keeping enough coordinates to preserve that structure introduces fractional participation but does not gain the high-multiplicity compression the programme is meant to study. That is a reason the coding results provide little support for a continuous mirror, not a proof that every mass-based model fails.
So the negative case is strongest at the anchor gate: none of the named results establishes the kind of computational result ChoCo is looking to continuize, and the proposed mass-based dimension problem is an adjacent research direction rather than a mirror of one. I cannot honestly establish the broader claim that no scenario could make such a direction worthwhile; a continuous representation theory for mass-based voting may be interesting if the authors recognize 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.