AFTD Auto-Formalizing Theoretical Domains

pcy_fixed_point_of_root

lemmaverified

For y > 0 the venue-0 impact of the three-type instance is positive, and a root of pcy_poly gives the fixed-point condition u_1 = y u_0.

Statement

open Finset in
lemma pcy_fixed_point_of_root (y : ℝ) (hy : 0 < y) :
    0 < pcy_impact y 0 ∧ (pcy_poly y = 0 → pcy_impact y 1 = y * pcy_impact y 0)
Source
step towards publication_choice_three_equilibria / publication_choice_three_types_three_equilibria (The Publication Choice Problem, AAAI 2026 (arXiv:2511.13678v2), Theorem 4.1 and Conjecture 4.2)
Verified
06 Oct 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 7
  • pcp_venue_impactVenue impact (eq. (2)): the impact of venue j is the average type of the researchers publishing there, weighted by type densities and…
  • pcp_half_two_actionThe closed-form action profile of Lemma 3.1 for alpha = 1/2, beta = 2: type i publishes v_j^4 / (c_ij^2 sum_l v_l^4 / c_il) at venue j.
  • pcy_thetaTypes of the three-type instance: theta = (1, 2, 8).
  • pcy_muType densities of the three-type instance: (3/10, 1/20, 13/20).
  • pcy_costCosts of the three-type instance: venue 0 costs 1 for every type; venue 1 costs 128, 64 and 8.
  • pcy_impactVenue impacts induced in the three-type instance by best responses to the normalised impacts (1, y).
  • pcy_polyThe cleared fixed-point polynomial A B - y C E of the three-type instance.
Used by 1

Lean source view module on GitHub

open Finset in/-- For `y > 0`, the impact of venue 0 is positive, and a root of `pcy_poly` gives the fixed-point condition `pcy_impact y 1 = y * pcy_impact y 0`. -/lemma pcy_fixed_point_of_root (y : ℝ) (hy : 0 < y) :    0 < pcy_impact y 0 ∧ (pcy_poly y = 0 → pcy_impact y 1 = y * pcy_impact y 0) := by  have h1 : (0 : ℝ) < 128 + y ^ 4 := by positivity  have h2 : (0 : ℝ) < 64 + y ^ 4 := by positivity  have h3 : (0 : ℝ) < 8 + y ^ 4 := by positivity  have hy4 : (0 : ℝ) < y ^ 4 := by positivity  simp only [pcy_impact, pcp_venue_impact, pcp_half_two_action, Fin.sum_univ_three, Fin.sum_univ_two,    pcy_cost, pcy_mu, pcy_theta]  simp  constructor  · positivity  · intro hP    have hE : (0 : ℝ) < (3 / 10 / 128) * ((64 + y ^ 4) * (8 + y ^ 4)) + (1 / 20 / 64) * ((128 + y ^ 4) * (8 + y ^ 4)) +      (13 / 20 / 8) * ((128 + y ^ 4) * (64 + y ^ 4)) := by positivity    have hB : (0 : ℝ) < (3 / 10 * 128) * ((64 + y ^ 4) * (8 + y ^ 4)) + (1 / 20 * 64) * ((128 + y ^ 4) * (8 + y ^ 4)) +      (13 / 20 * 8) * ((128 + y ^ 4) * (64 + y ^ 4)) := by positivity    rw [← sub_eq_zero]    rw [show pcy_poly y = 0 ↔ pcy_poly y / (((3 / 10 / 128) * ((64 + y ^ 4) * (8 + y ^ 4)) +        (1 / 20 / 64) * ((128 + y ^ 4) * (8 + y ^ 4)) + (13 / 20 / 8) * ((128 + y ^ 4) * (64 + y ^ 4))) *        ((3 / 10 * 128) * ((64 + y ^ 4) * (8 + y ^ 4)) + (1 / 20 * 64) * ((128 + y ^ 4) * (8 + y ^ 4)) +        (13 / 20 * 8) * ((128 + y ^ 4) * (64 + y ^ 4)))) = 0 by          rw [div_eq_zero_iff]; constructor          · exact Or.inl          · rintro (h | h)            · exact h            · exact absurd h (mul_pos hE hB).ne'] at hP    rw [← hP, pcy_poly]    field_simp
See it in the graphDiscuss this resultChallenge it