Theorem 1: uniqueness of the canonical core under two-sided Clifford equivalence (seed result)
OpenQAlgorithms.Papers.Independent.Intakenext100.P82.canonical_core_uniqueAll circuits are single-qubit circuits without ancillas or measurements. Products act from right to left. Let , , , and . The two distinct syllable letters are and . A normalized word is empty, or ends in with no adjacent letters. A canonical word is normalized and its first four syllables, when present, are all . Write for the number of letters in ; Clifford gates have zero charged cost.
Let be canonical syllable words. If there are with as projective gates, then
as words in the two-letter alphabet. This gives a unique canonical core for each Clifford double coset. It makes no uniqueness assertion about the surrounding Clifford gates.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P82
noncomputable section
namespace QAlgorithms.Papers.Independent.Intakenext100.P82
open scoped BigOperators
local notation "Mat" => Matrix Bool Bool ℂ
/-- Theorem 1, PDF p.3, grammar p.2, proof pp.9--10.
Only the canonical syllable cores are syntactically equal. -/
theorem canonical_core_unique (c1 c2 : List Syllable)
(h1 : BsCanonical c1) (h2 : BsCanonical c2)
(heq : ∃ g1 g2 : Fin 24,
SameGate (syllableValue c2) (cliffordMatrix g1 * syllableValue c1 * cliffordMatrix g2)) :
c1 = c2 := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P82
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be any finite lists of syllables , where these syllables represent , respectively, for
Let multiply the syllable matrices in list order, with empty product . Suppose each is empty or ends in and has no adjacent syllables, and suppose every one of its first syllables is . Define , for , by
If there exist indices and a complex number with such that
then as lists, including equality of their lengths and every syllable. The hypotheses allow either list to be empty; for a list shorter than four, they require every syllable to be .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.