| paper | d4-conf-atal-2013-0170 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | yes |
A dispatch LP over household types is a coherent continuous-population question, and both cases acknowledge that. But the paper states no numbered Theorem, Lemma, Corollary, or Proposition establishing a computational result: its LPs appear as equations used in experiments, while the polynomial-time Shapley approximation is attributed to Castro et al. The named-result requirement therefore fails, even though the proposed mirror is plausible.
fails bit a — no named computational result to mirror
The proposed dispatch mirror would cover the community battery-charging optimization in Eq. \(3\) and its individual-rationality constraints. It leaves the Shapley allocation in Eq. \(6\), the cited sampling method, and the empirical performance claims without a mirror.
The strongest positive mirror is the paper’s community dispatch problem. A plausible high-multiplicity setting is a remote village with hundreds of households but relatively few repeated combinations of solar or wind generation, load profile, and battery specifications. A type would include the full per-home data used by the model—generation and load profiles, battery capacity, charge and discharge limits, and efficiency. Its mass \(\mu_t\) is the fraction of households of that type.
A precise candidate is Continuous Community Exchange Dispatch. Given a finite list of types with rational masses and a shared time horizon, choose per-household generation, allocation, battery charge and discharge, battery state, and link-flow schedules for each type. Minimize \(\sum_t \mu_t\sum_i c_{ti}\), subject to the paper’s physical constraints, balanced community flows, each type retaining its disconnected utility, and each type’s battery use not exceeding its disconnected baseline. Return an optimal schedule and its total charging. The paper’s model is linear, so this type-level version is an LP with size polynomial in the support \(\tau\), horizon, and input bit length; I would expect Class A. This mirrors the population and energy-exchange decision directly. It covers the dispatch problem, not the paper’s Shapley fairness calculation or its empirical percentage reductions.
There is a qualification even to that formulation: the prose says connected battery use will not increase, but displayed constraint \(o12\) has the inequality in the opposite direction. The candidate above follows the stated intent; a faithful formal mirror would need that discrepancy resolved.
Under your anchor rule, though, this is only a plausible mirror, not an anchored one. The paper contains no numbered Theorem, Lemma, Corollary, or Proposition asserting a computational result. Its LP and Shapley definitions are equations; the polynomial-time sampling claim relies on Castro et al.’s algorithm, and the paper’s own scalability and savings claims are empirical. So I cannot honestly name a qualifying anchor or claim that this candidate mirrors one. The main weakness is that the paper’s most distinctive computation is Shapley allocation among finite agents: extending that to a continuum is not automatic, and treating each type as a single player would change the game.
The strongest case against this paper as a ChoCo anchor is that it states no numbered computational result. Equation (3) defines the community optimization problem; equation (6) defines its Shapley allocation; and the paper reports runtimes and savings empirically. Its polynomial-time Shapley approximation relies on Castro et al.’s algorithm. None is a theorem establishing the complexity of a problem from this paper. Under the requested anchor rule, the proponent’s community-dispatch LP is a newly framed question, not a mirror of a qualifying result.
That does not make the dispatch question incoherent. A type-level version with repeated household profiles is plausible, and the paper’s \(o12\) inequality can be corrected to match its stated intent. But I cannot use “the LP is already tractable” or “continuization changes nothing” to dismiss it: the programme explicitly treats a dull answer as a valid result. The defensible objection is narrower: the paper did not establish a computational claim for that LP to anchor the proposed mirror.
The best fairness mirror is also hard to dismiss outright. One could extend the coalition value from finite subsets to coalition masses and ask whether per-household Shapley allocations converge to a continuous allocation, such as an Aumann–Shapley value, and how to compute it. That would preserve the paper’s concern with fair division more closely than treating each type as a player. But it is a new limit and complexity question, not a result the paper proves.
So the negative case is strong on the assignment’s anchor test: no qualifying result was identified. It is weak as a universal claim that no worthwhile continuous mirror exists. The dispatch LP is a sensible candidate, and a continuous fairness question may be too. I would argue against grading this paper as an anchored computational contribution, but I cannot honestly rule out all useful mirrors.
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.