is_bounded_bias_rank_one_product
A function R : (Fin n → Bool) → ℝ on the n-dimensional Boolean cube is a bounded-bias rank-one product (is_bounded_bias_rank_one_product R) if there exists a family of nonnegative one-variable functions g : Fin n → Bool → ℝ such that for every coordinate i : Fin n, either g i true = 0, g i false = 0, or g i true = g i false, and R τ = ∏ i : Fin n, g i (τ i) for all truth assignments τ : Fin n → Bool.
Statement
def is_bounded_bias_rank_one_product {n : ℕ} (R : (Fin n → Bool) → ℝ) : Prop
- Defined in
- arXiv:2611.01055, Definition 2.5 and Lemma 2.6
- Verified
- 04 Nov 2026
- Axioms
- none listed
Read back from the Lean
Given a natural number n and a function R : (Fin n → Bool) → ℝ, is_bounded_bias_rank_one_product R asserts that there exists a function g : Fin n → Bool → ℝ such that:
1. For all i : Fin n and all b : Bool, 0 ≤ g i b.
2. For all i : Fin n, g i true = 0 ∨ g i false = 0 ∨ g i true = g i false (that is, for each coordinate, at least one of the values is zero, or both values are equal).
3. For all τ : Fin n → Bool, R τ equals the product over all i : Fin n of g i (τ 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
/-- A function R on the n-dimensional Boolean cube is a bounded-bias rank-one product if it factors as a nonnegative product whose one-variable factors have biases in {0, 1/2, 1}. -/ def is_bounded_bias_rank_one_product {n : ℕ} (R : (Fin n → Bool) → ℝ) : Prop := ∃ g : Fin n → Bool → ℝ, (∀ i b, 0 ≤ g i b) ∧ (∀ i, g i true = 0 ∨ g i false = 0 ∨ g i true = g i false) ∧ ∀ τ, R τ = ∏ i : Fin n, g i (τ i)