Lean, informalized

Nielsen & Chuang · Chapter 10, Quantum error-correction · p. 464

Exercise 10.42 - After both measurements, the correction leaves the receiving qubit in the input's eigenstate

Stabilizers check teleportation of XX and ZZ eigenstates.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Exercise

Alice wants to send Bob the state of a qubit, qubit 0, and the two share a pair of qubits, 1 and 2, in the Bell state (|00⟩+|11⟩)/2(|00\rangle + |11\rangle)/\sqrt{2}, qubit 2 on Bob’s side. Alice applies a controlled NOT gate from qubit 0 to qubit 1 and a Hadamard gate to qubit 0, measures qubits 0 and 1 in the computational basis with outcomes m0m_0 and m1m_1, and sends Bob the two bits. Bob corrects qubit 2 with Pauli operators chosen by the bits, and qubit 2 then holds the state Alice started with.

The exercise is to verify this with stabilizers: a state is followed through the circuit by the Pauli operators that fix it, where a Pauli operator PP fixes a vector |ψ⟩|\psi\rangle when P|ψ⟩=|ψ⟩P|\psi\rangle = |\psi\rangle. The inputs are eigenstates of the Pauli operators XX and ZZ: the state |+⟩|+\rangle for b=0b = 0 or |−⟩|-\rangle for b=1b = 1, which XX multiplies by (−1)b(-1)^b, and the state |a⟩|a\rangle for a=0a = 0 or 11, which ZZ multiplies by (−1)a(-1)^a. Write XkX_k and ZkZ_k for XX and ZZ acting on qubit kk, with the identity on the other qubits.

The library states what Bob’s correction makes of the measured state, for each of the two inputs.

Pauli Operators

The Pauli matrices σ0,…,σ3\sigma_0, \dots, \sigma_3 are the four complex 2×22 \times 2 matrices σ0=I\sigma_0 = I, the identity, and σ1=X\sigma_1 = X, σ2=Y\sigma_2 = Y, σ3=Z\sigma_3 = Z, which in the computational basis |0⟩,|1⟩|0\rangle, |1\rangle are X=(0110),Y=(0−ii0),Z=(100−1).X = \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}, \qquad Y = \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix}, \qquad Z = \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix}.(1)✦

A Pauli string on nn qubits is the operator σg0⊗⋯⊗σgn−1\sigma_{g_0} \otimes \cdots \otimes \sigma_{g_{n-1}} that acts on each qubit kk with one Pauli matrix σgk\sigma_{g_k}, chosen by an index gk∈{0,1,2,3}g_k \in \{0, 1, 2, 3\}. Its matrix element between the bit strings s=s0⋯sn−1s = s_0 \cdots s_{n-1} and t=t0⋯tn−1t = t_0 \cdots t_{n-1} is the product of single qubit entries, ⟨s|σg0⊗⋯⊗σgn−1|t⟩=∏k=0n−1(σgk)sktk.\langle s |\, \sigma_{g_0} \otimes \cdots \otimes \sigma_{g_{n-1}} \,| t \rangle = \prod_{k=0}^{n-1} (\sigma_{g_k})_{s_k t_k}.(2)✦ The operators XkX_k and ZkZ_k on three qubits and their products, such as Z0X2=Z⊗I⊗XZ_0X_2 = Z \otimes I \otimes X and Z1Z2=I⊗Z⊗ZZ_1Z_2 = I \otimes Z \otimes Z, are Pauli strings.

Bob’s correction for the outcomes m0m_0 and m1m_1 is the operator Cm0m1C_{m_0m_1} on the three qubits that applies Xm1X^{m_1} and then Zm0Z^{m_0} to qubit 2: Cm0m1=Z2m0X2m1.C_{m_0m_1} = Z_2^{m_0} X_2^{m_1}.(3)✦

The Claim

Theorem 1. Let m0,m1,a,b∈{0,1}m_0, m_1, a, b \in \{0, 1\} and let |ψ⟩|\psi\rangle be a vector in the state space (ℂ2)⊗3(\mathbb{C}^2)^{\otimes 3} of three qubits. If (−1)m0Z0(-1)^{m_0}Z_0 and (−1)bZ0X2(-1)^bZ_0X_2 fix |ψ⟩|\psi\rangle, then X2Cm0m1|ψ⟩=(−1)bCm0m1|ψ⟩.X_2\, C_{m_0m_1}|\psi\rangle = (-1)^b\, C_{m_0m_1}|\psi\rangle.(4)✦ If (−1)m1Z1(-1)^{m_1}Z_1 and (−1)aZ1Z2(-1)^aZ_1Z_2 fix |ψ⟩|\psi\rangle, then Z2Cm0m1|ψ⟩=(−1)aCm0m1|ψ⟩.Z_2\, C_{m_0m_1}|\psi\rangle = (-1)^a\, C_{m_0m_1}|\psi\rangle.(5)✦

What Is Checked

The claim is stated over six Pauli strings on three qubits, Z0Z_0, Z1Z_1, Z0X2Z_0X_2, Z1Z2Z_1Z_2, X2X_2 and Z2Z_2, each built by (2) from the Pauli matrices (1), and over Bob’s correction (3). About these operators the library proves the two parts of the theorem, (4) for the input in the XX basis and (5) for the input in the ZZ basis.