No-cloning principle
ProvedQAlgorithms.Nannicini.no_cloningLet . There is no unitary matrix acting on two -qubit registers, i.e. on , such that for every -qubit quantum state (unit vector)
A single unitary cannot copy every quantum state; in particular the destructive effect of measurement cannot be avoided by cloning the state beforehand.
Formalization Note The two registers are indexed by pairs of bit strings, first register first; this is the source's with the first qubits forming the first register. Equality is exact, as in the source. The statement also holds at , so no positivity hypothesis on is added.
import Definitions.Def_QAlgorithms_Defs_Measurement set_option autoImplicit false
namespace QAlgorithms.Nannicini
/-- Nannicini p.23, Theorem 1.32 (No-cloning principle): there is no unitary matrix on two
`q`-qubit registers that maps `|ψ⟩_q |0⟩_q` to `|ψ⟩_q |ψ⟩_q` for every `q`-qubit quantum state
`|ψ⟩`. -/
theorem no_cloning (q : ℕ) :
¬ ∃ U : Matrix (Qubits q × Qubits q) (Qubits q × Qubits q) ℂ,
U ∈ Matrix.unitaryGroup (Qubits q × Qubits q) ℂ ∧
∀ ψ : EuclideanSpace ℂ (Qubits q), IsState ψ →
act U (tensorVec ψ (zeroKet q)) = tensorVec ψ ψ := by
sorry
end QAlgorithms.Nannicini
Read-back
What the Lean code literally says, in plain math · claude-opus-5-5
For every natural number , including , there is no complex matrix on the two-register space with both of the following properties:
- is unitary, i.e. ;
- for every unit vector (with ),
Here is the standard basis vector at the all-zeros string. For vectors , the product is the vector with entries , the first register being the first factor.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.