Documentation

LeanPool.ErdosGinzburgZiv.EGZ.Expansion.Subweights

Selecting submultisets with prescribed fibre masses #

All selections are made in natural multiplicities, so coincident vector values are still allowed to occupy distinct positions.

theorem EGZ.Expansion.exists_subweight_of_mass_le {A : Type u_1} [Fintype A] (w : A → ℕ) {n : ℕ} (hn : n ≤ natMass w) :
∃ u ≤ w, natMass u = n
theorem EGZ.Expansion.exists_subweight_pushWeight {A : Type u_1} {B : Type u_2} [Fintype A] (π : A → B) (w : A → ℕ) (a : B → ℕ) (ha : a ≤ pushWeight π w) :
∃ u ≤ w, pushWeight π u = a

Every collection of fibre counts dominated by the available counts can be realized by a submultiset.

theorem EGZ.Expansion.pushWeight_add {A : Type u_1} {B : Type u_2} [Fintype A] (π : A → B) (u w : A → ℕ) :
pushWeight π (u + w) = pushWeight π u + pushWeight π w
theorem EGZ.Expansion.pushWeight_sub {A : Type u_1} {B : Type u_2} [Fintype A] (π : A → B) {u w : A → ℕ} (hu : u ≤ w) :
pushWeight π (w - u) = pushWeight π w - pushWeight π u
theorem EGZ.Expansion.pushWeight_sum {A : Type u_1} {B : Type u_2} {I : Type u_3} [Fintype A] (π : A → B) (s : Finset I) (w : I → A → ℕ) :
pushWeight π (∑ i ∈ s, w i) = ∑ i ∈ s, pushWeight π (w i)