AFTD Auto-Formalizing Theoretical Domains

predecessors

def

Given a finite type α with decidable equality, an ordering bijection e : Fin (Fintype.card α) ≃ α mapping indices to players, and a player i : α, predecessors e i is the finite set of players e(j) for all indices j : Fin (Fintype.card α) strictly less than e.symm i.

Statement

def predecessors {α : Type*} [DecidableEq α] [Fintype α] (e : Fin (Fintype.card α) ≃ α) (i : α) : Finset α
Defined in
arXiv:2611.16207, Section 2
Verified
13 Nov 2026
Axioms
none listed
Used by 3
  • marginal_vectorFor a cooperative game v : Finset α → ℝ on a finite type α with decidable equality and an ordering bijection e : Fin (Fintype.card α) ≃ α,…
  • local_no_swap_of_in_coreFor any cooperative game v on a finite type α satisfying v(∅) = 0 and any ordering bijection e : Fin (Fintype.card α) ≃ α, if the marginal…
  • two_thirds_convex_three_player_noncore_ordering_fraction_ge_two_thirdsThere exists a 2/3-convex cooperative game v on 3 players (Fin 3) satisfying v(∅) = 0 for which the probability that a uniformly random…

Read back from the Lean

For a finite type α with decidable equality, an equivalence e : Fin (Fintype.card α) ≃ α, and an element i : α, predecessors e i is defined to be the finite set (Finset α) obtained by filtering the universal set of Fin (Fintype.card α) to those indices j strictly less than e.symm i (under the standard order on Fin), and then mapping that filtered set through e. In other words, it is the set of all elements in α whose index under e is strictly less than the index of i.

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

/-- The set of predecessors of player i under ordering e. -/
def predecessors {α : Type*} [DecidableEq α] [Fintype α] (e : Fin (Fintype.card α) ≃ α) (i : α) : Finset α :=
  (Finset.univ.filter (fun (j : Fin (Fintype.card α)) => j < e.symm i)).image e
See it in the graphDiscuss this resultChallenge it