is_neutral_rule
Given types Agent and Object with decidable equality on Object, a rule f : (Agent -> LinearOrder Object) -> Agent -> Finset Object satisfies is_neutral_rule if for every profile succ : Agent -> LinearOrder Object, every permutation σ : Equiv.Perm Object, and every agent i : Agent, the finite bundle f (fun j => @Equiv.linearOrder Object Object σ.symm (succ j) _) i equals the image under σ of the finite bundle f succ i, namely (f succ i).map σ.toEmbedding.
Statement
def is_neutral_rule {Agent Object : Type*} [DecidableEq Object] (f : (Agent → LinearOrder Object) → Agent → Finset Object) : Prop
- Defined in
- arXiv:2611.09685 Section 2.3
- Verified
- 09 Nov 2026
- Axioms
- none listed
Read back from the Lean
A rule f, which assigns to each preference profile (a function assigning a LinearOrder on Object to each Agent) and each Agent a Finset Object, satisfies is_neutral_rule if and only if:
For every profile succ : Agent → LinearOrder Object,
for every permutation σ : Equiv.Perm Object of Object,
and for every agent i : Agent,
the finset of objects assigned to i under the profile where each agent j has their linear order pulled back by σ.symm (so that $x \le y$ in agent $j$'s new order if and only if $\sigma^{-1}(x) \le \sigma^{-1}(y)$ in succ j)
is equal to the image under σ (via σ.toEmbedding) of the finset of objects assigned to i under the original profile succ.
Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.
Lean source view module on GitHub
/-- A rule is neutral if permuting objects permutes the assigned bundles accordingly. -/ def is_neutral_rule {Agent Object : Type*} [DecidableEq Object] (f : (Agent → LinearOrder Object) → Agent → Finset Object) : Prop := ∀ (succ : Agent → LinearOrder Object) (σ : Equiv.Perm Object) (i : Agent), f (fun j => @Equiv.linearOrder Object Object σ.symm (succ j) _) i = (f succ i).map σ.toEmbedding