Eq. (1.1) = (3.21), pp. 2, 15 — every isotropic cloner obeys , and every point of the ellipse is attained
ProvedCerfCloning.Isotropic.no_cloning_inequalityLet . A isotropic cloner of an -level system with depolarizing fractions is a quantum channel (completely positive and trace preserving) from operators on the -level input to operators on , two -level outputs, whose two outputs emerge from depolarizing channels: for every operator ,
i.e. copy is the input replaced by a random state with probability , and copy with probability . Write and .
- No-cloning inequality. Every isotropic cloner satisfies
- Tightness. For all with there is an isotropic cloner with depolarizing fractions and .
So the region of inside the ellipse , which would represent close-to-perfect cloning, is forbidden by quantum mechanics, and the ellipse itself is the exact boundary of what isotropic cloners achieve (the paper attains it with isotropic Heisenberg cloning machines). For this is the qubit no-cloning inequality; as the ellipse tends to the line .
Formalization Note A channel is a finite family of Kraus operators with (QAlgorithms.IsChannel), which in finite dimension is exactly a completely positive trace-preserving map. The first factor of the output space is . The output conditions are imposed on every matrix ; for a linear map that is the same as imposing them on all density matrices, or on all pure states. No sign condition is put on in clause 1: complete positivity of the output channels forces , so Real.sqrt returns the paper's and . For the fractions are determined by the channel.
Differences from the source:
- encoding: an -dimensional quantum state, unrestricted → hypothesis ; reason: at the depolarizing fraction is not determined by the channel (every fraction gives the identity map), so the identity channel would be an isotropic cloner with ; the paper never considers , where there is nothing to clone.
- reading: "the domain … inside a particular ellipse … is forbidden" for "all isotropic cloning machines", "cloners whose outputs emerge from (distinct) depolarizing channels" (abstract, p. 2) → clause 1 for every quantum channel whose two outputs are depolarizing; reason: this is the paper's own definition of an isotropic cloner and its claim on p. 2; the derivation in §III C runs inside the Heisenberg cloning machines, but the claim is stated for all isotropic cloners, and the narrower class would make the inequality a restatement of the ellipse (3.20).
- reading: "It is a tight inequality … which is saturated with the HCMs" → clause 2: every point of the ellipse with is attained by some isotropic cloner; reason: tightness and saturation are existence claims; the paper's witness for the point , , is the HCM (3.8) with , which the statement does not name.
- generalization: , with real and positive → via
Real.sqrt, and the hypotheses in clause 2; reason: the same numbers for ; clause 2 also covers the two endpoints and of the ellipse, so it contains the paper's claim; in clause 1 nonnegativity follows from complete positivity. - encoding: outputs depolarizing on input states → on every matrix ; reason: a channel is linear and density matrices span all matrices, so the conditions are equivalent.
import Mathlib import Definitions.Def_CerfCloning_Isotropic_Model open scoped Matrix ComplexOrder BigOperators open QAlgorithms
namespace CerfCloning.Isotropic
theorem no_cloning_inequality (N : ℕ) [NeZero N] (hN : 2 ≤ N) :
(∀ (T : Superop (Fin N) (Fin N × Fin N)) (pa pb : ℝ), IsIsotropicCloner N T pa pb →
1 ≤ pa + 2 / (N : ℝ) * (Real.sqrt pa * Real.sqrt pb) + pb) ∧
(∀ pa pb : ℝ, 0 ≤ pa → 0 ≤ pb → pa + 2 / (N : ℝ) * (Real.sqrt pa * Real.sqrt pb) + pb = 1 →
∃ T : Superop (Fin N) (Fin N × Fin N), IsIsotropicCloner N T pa pb) := by sorry
end CerfCloning.Isotropic
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.