AFTD Auto-Formalizing Theoretical Domains

code_bad_projective_set

def

Given an m by n generator matrix G over a field F and a distance parameter d, the bad projective set consists of all projective points whose codeword under G has Hamming weight at most d - 1.

Statement

def code_bad_projective_set {F : Type*} [Field F] [DecidableEq F] {m n : ℕ}
    (G : Matrix (Fin m) (Fin n) F) (d : ℕ) : Set (Projectivization F (Fin m → F))
Defined in
arXiv:2611.16447 Section 1.1, Lemma 3.8
Verified
14 Nov 2026
Axioms
none listed

Read back from the Lean

For a field F with decidable equality, natural numbers m and n, an m × n matrix G over F, and a natural number d, code_bad_projective_set G d is the set of points p in the projectivization of Fin m → F (Projectivization F (Fin m → F)) such that the Hamming norm of the vector-matrix product p.rep ᵥ* G is at most d - 1 (where p.rep is the chosen representative vector of p).

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

/-- The set S(G) of projective directions whose codeword under G has Hamming weight at most d - 1. -/
def code_bad_projective_set {F : Type*} [Field F] [DecidableEq F] {m n : ℕ}
    (G : Matrix (Fin m) (Fin n) F) (d : ℕ) : Set (Projectivization F (Fin m → F)) := { p | hammingNorm (Matrix.vecMul p.rep G) ≤ d - 1 }
See it in the graphDiscuss this resultChallenge it