| paper | d4-conf-ijcai-2016-0283 |
| 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 3 give impossibility claims about mechanisms, while Theorem 2 proves the axiomatic properties of an explicitly specified mechanism. The paper states no computational result, so bit (a) fails; the opponent also shows that a nontrivial population computation would require a new task with a changed unit of agency.
fails bit a — no named computational result to mirror
No result is covered by a continuous computational mirror; Theorem 2 gives a mechanism property, while Theorems 1 and 3 give mechanism impossibility claims.
This paper has no named computational result to anchor a continuous mirror to. Its numbered results are proved here, but they concern mechanism design: Theorem 1 says that with two agents and at least three alternatives, every mechanism satisfying A1–A4 on the general utility space \(U_0\) is dictatorial; Theorem 2 constructs a mechanism satisfying A1–A7 on parallel domains; and Theorem 3 rules out such a mechanism for type spaces containing \(U_{\mathrm{QL}}\) but not contained in the largest parallel domain \(U_k\). None asserts a complexity bound for a computational problem.
Theorem 2 is the closest foothold: it gives an explicit VCG-like mechanism. But the paper specifies neither a finite encoding of its arbitrary utility functions nor a running-time claim, so calling it “in P” would add assumptions and a result of our own. The cited computational-complexity papers are background, not results of this paper.
So there are no eligible anchors, and I can’t honestly provide the requested continuous computational problem per anchor. A continuum-population extension of Theorem 2 might be a separate research question, but it would not make the positive case this prompt asks for.
The appended case offers no continuous mirror or named computational anchor; it says Theorems 1–3 are mechanism-design results, not complexity results. That is the strongest case against a mirror of this paper’s contribution. None states a running time or studies an algorithmic problem, and the paper gives no finite encoding for its arbitrary utility functions. Theorem 2’s explicit mechanism does not, by itself, supply a complexity result.
The best computational extension would give a society as a finite-support distribution over complete utility types and ask for the parallel-domain mechanism’s outcome. Given each type’s willingness-to-pay values \(p_{t,j}\), that reduces to computing \(\arg\max_j \sum_t \mu_t p_{t,j}\), a direct \(O(\tau m)\) aggregation. With arbitrary utility functions, the effort lies in encoding and evaluating those functions, not in population continuization. And in an atomless population, an individual’s report has zero effect on the outcome: the individual externality payment vanishes, so the paper’s dominant-strategy incentive condition loses its force. Treating a positive mass of people as one strategic bloc could restore an effect, but that changes the unit of agency and introduces a coalition or control problem the paper does not define.
A high-multiplicity setting for collective decisions is plausible, so this does not prove that no adjacent research question could be worthwhile. It does make the negative case strong for this paper: to get a nontrivial computational question, one must add a task such as mass manipulation, specify costs and utility encodings, and decide what counts as an agent. Those additions may be valuable, but they are new computational problems rather than continuous mirrors of the paper’s 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.