Wall: nonzero rational multiples preserve normality
OpenBorelNormality.isNormalReal_mul_ratWall's invariance theorem (1949). Let be an integer and a real number normal in base . If is a nonzero rational number, then is again normal in base :
Normality in a fixed base is thus invariant under nonzero rational scaling (and, applying the theorem to , the converse holds as well). This lets reductions renormalise a number — for instance scale it into — without losing normality, and is a basic structural lemma of the theory.
Formalization Note. The base and the number are implicit binders; is coerced from to . The predicate is Normal.IsNormalReal from Normal_Core.
import Mathlib import Definitions.Def_Normal_Core set_option autoImplicit false
namespace BorelNormality
theorem isNormalReal_mul_rat {b : ℕ} (hb : 2 ≤ b) {x : ℝ}
(hx : Normal.IsNormalReal b x) (a : ℚ) (ha : a ≠ 0) :
Normal.IsNormalReal b (x * (a : ℝ)) := by sorry
end BorelNormalityRead-back
What the Lean code literally says, in plain math · GLM (Z.ai), same-agent blind pass — not an independent sub-agent
For every natural number with , every real number , and every rational number : if Normal.IsNormalReal b x holds and , then Normal.IsNormalReal b (x \cdot a) holds, where is coerced to a real. Unfolding the predicate on both sides: if every word over occurs in the digits with limiting frequency , then the same holds for the digits of . The binders and are implicit; and are explicit hypotheses. No sign or size restriction is placed on or beyond ; digits are fractional-part digits in the canonical expansion.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.