persuasion_slice_ratio
theoremverifiedoriginal
The slice prior attains ratio (3m+1)/(2m+2). For the prior uniform over the 2m+1 states of persuasion_slice_prior m (coordinates = m-subsets of a (2m+1)-set), R_max / R_0 = (3m+1)/(2m+2) = 3/2 - 1/(m+1). For m = 4 this is 13/10 > 39/31.
Statement
open Finset in theorem persuasion_slice_ratio (m : ℕ) (hm : 1 ≤ m) : 2 * (m + 1) * persuasion_max_receiver_utility (persuasion_slice_prior m) = (3 * m + 1) * persuasion_prior_utility (persuasion_slice_prior m) ∧ 0 < persuasion_prior_utility (persuasion_slice_prior m)
- Related work
- operator (Claude), 2026-10-03: the optimal constant in the receiver-utility bound of arXiv:2606.22226 v1 (Sec. 5, first open direction: the constant lies between 39/31 and 3/2) is 3/2. The slice prior (2m+1 equally likely states, one coordinate per m-subset) has R_max/R_0 = (3m+1)/(2m+2)
- Verified
- 03 Oct 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 4
- persuasion_slice_priorThe slice prior: uniform over the 2m+1 states persuasion_slice_state m s.
- persuasion_max_receiver_utilityR_max(μ): the largest receiver utility among sender-optimal obedient schemes.
- persuasion_prior_utilityThe receiver's prior-only utility R_0(μ) = ∑_i max(p_i, 1 - p_i).
- persuasion_slice_valuesR_max and R_0 for the slice prior.
- Used by 1
- persuasion_receiver_ratio_sharpThe 3/2 bound of arXiv:2606.22226 is sharp. For every c < 3/2 there is a prior with R_max(μ) > c · R_0(μ).
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
open Finset in/-- `(m+1) C(2m+1, m) = 2 (2m+1) C(2m-1, m)` for `m ≥ 1`. -/lemma persuasion_slice_choose_identity (m : ℕ) (hm : 1 ≤ m) : (m + 1) * (2 * m + 1).choose m = 2 * (2 * m + 1) * (2 * m - 1).choose m := by obtain ⟨k, rfl⟩ : ∃ k, m = k + 1 := ⟨m - 1, by omega⟩ have c1 : (2 * (k + 1)).choose (k + 1) = 2 * (2 * (k + 1) - 1).choose (k + 1) := by rw [show 2 * (k + 1) = (2 * k + 1) + 1 by ring, Nat.choose_succ_succ, show 2 * k + 1 + 1 - 1 = 2 * k + 1 by omega] have : (2 * k + 1).choose k = (2 * k + 1).choose (k + 1) := by rw [← Nat.choose_symm (by omega : k ≤ 2 * k + 1)]; congr 1; omega rw [this]; ring have h := Nat.add_one_mul_choose_eq (2 * (k + 1)) (k + 1) have hs : (2 * (k + 1) + 1).choose (k + 1 + 1) = (2 * (k + 1) + 1).choose (k + 1) := by rw [← Nat.choose_symm (by omega : k + 1 + 1 ≤ 2 * (k + 1) + 1)]; congr 1; omega rw [hs, c1] at h linarithopen Finset in/-- **The slice prior attains ratio `(3m+1)/(2m+2)`.** For the prior uniform over the `2m+1`states of `persuasion_slice_prior m` (coordinates = `m`-subsets of a `(2m+1)`-set),`R_max / R_0 = (3m+1)/(2m+2) = 3/2 - 1/(m+1)`. For `m = 4` this is `13/10 > 39/31`. -/theorem persuasion_slice_ratio (m : ℕ) (hm : 1 ≤ m) : 2 * (m + 1) * persuasion_max_receiver_utility (persuasion_slice_prior m) = (3 * m + 1) * persuasion_prior_utility (persuasion_slice_prior m) ∧ 0 < persuasion_prior_utility (persuasion_slice_prior m) := by obtain ⟨h1, h2⟩ := persuasion_slice_values m hm have hid : ((m : ℝ) + 1) * (2 * m + 1).choose m = 2 * (2 * m + 1) * (2 * m - 1).choose m := by exact_mod_cast persuasion_slice_choose_identity m hm have hK : (0 : ℝ) < 2 * m + 1 := by positivity have hn : (0 : ℝ) < (2 * m + 1).choose m := by exact_mod_cast Nat.choose_pos (by omega) rw [h1, h2] refine ⟨?_, by positivity⟩ field_simp nlinarith [hid]