AFTD Auto-Formalizing Theoretical Domains

plackett_luce_prob

def

Plackett-Luce probability of a ranking (best first): the product over positions of the weight of the alternative placed there divided by the total weight of it and all alternatives ranked below it.

Statement

noncomputable def plackett_luce_prob (w : ℕ → ℝ) (l : List ℕ) : ℝ
Defined in
Preference Elicitation as Average-Case Sorting, AAAI 2021, Sec. 5 (Plackett-Luce model D_w and its entropy H(D_w)); formalized by the operator (Claude), 2026-10-03
Verified
03 Oct 2026
Axioms
none listed
Used by 5
  • plackett_luce_entropy_geometric_leFor geometric weights phi^a with 0 < phi < 1, the Plackett-Luce entropy on any set of n alternatives is at most C(phi) * n with C(phi) =…
  • plackett_luce_surprisal_measureFor weights w : ℕ → ℝ and number of items n : ℕ, the pushforward measure on ℝ of the standardized surprisal (-log P(l) - H) / sqrt(n)…
  • sum_plackett_luce_probWith positive weights, the Plackett-Luce probabilities of the rankings of S sum to 1.
  • plackett_luce_prob_consIf a is in S and t ranks S minus a, the Plackett-Luce probability of a followed by t is w(a)/w(S) times that of t.
  • plackett_luce_entropyShannon entropy (natural logarithm) of the Plackett-Luce distribution over the rankings of S.

Lean source view module on GitHub

/-- Plackett–Luce probability of the ranking `l` (best first) under weights `w`: the product over positions of `w a / (w a + total weight of the alternatives below a)`. -/
noncomputable def plackett_luce_prob (w : ℕ → ℝ) (l : List ℕ) : ℝ := match l with
  | [] => 1
  | a :: t => w a / (w a + (t.map w).sum) * plackett_luce_prob w t
See it in the graphDiscuss this resultChallenge it