liveBook · Manning

Phase kickback

As explained in Quantum Programming in Depth

Phase kickback is a quantum effect in which a controlled operation transfers a phase associated with its target to the relative phase of the control-qubit state.

Phase kickback transfers a phase from a controlled unitary’s target to the control’s relative phase. Prepare the control in superposition and the target in an eigenstate: if U|ψ⟩ = e^(2πiθ)|ψ⟩, controlled-U maps |+⟩|ψ⟩ to (|0⟩ + e^(2πiθ)|1⟩)|ψ⟩/√2. The target is unchanged and unentangled; the control now carries θ. section 5.3.1 section 5.3.2

The effect is observable only after choosing a measurement that converts relative phase into differing outcome probabilities. For eigenvalues +1 and −1, the control remains |+⟩ or becomes |−⟩; a Hadamard and computational-basis measurement distinguish them. A generic superposition need not become orthogonal after a sign change, requiring repeated experiments. section 5.1.1 section 5.3.2

Figure 5.5 To use the phase kickback trick for phase estimation, we need a control qubit in a superposition state (the \(\ket{+}\) state is commonly used) and a target register of qubits in the state that is an eigenvector of the unitary we’re exploring. When we apply the controlled variant of this unitary, the target register remains unchanged, but the control qubit acquires a relative phase on the basis state \(\ket{1}\). This relative phase can be analyzed to extract informa-tion about the eigenphase that corresponds to the eigenvector.
figure

Inspect the essential circuit pattern: a |+⟩ control, an eigenstate target register, and controlled-U leave the target intact while encoding its eigenphase on the control. section 5.3.1

Phase kickback supplies a reusable bridge from controlled computation to phase-based algorithms:

  • Phase estimation encodes an eigenphase on one or more control qubits. Iterative methods estimate probabilities over repeated runs; adaptive methods recover binary digits with successively adjusted circuits; quantum phase estimation applies inverse quantum Fourier transform before measurement. section 5.3.2 section 5.3.3 section 5.3.5
  • A marking oracle can become a phase oracle by applying it to an auxiliary |−⟩ qubit. Marked inputs acquire a −1 relative phase while the auxiliary qubit remains unchanged. section 7.1.1 section 7.1.3

Turning a marking oracle into a phase oracle

For marked bit strings 001 and 110, prepare an input superposition and an auxiliary |−⟩ state, then apply the marking oracle. The input amplitudes for 001 and 110 change sign, while the auxiliary qubit can be uncomputed. Qiskit retains the auxiliary qubit as part of the circuit; Q# allocates, resets, and releases it. section 7.1.2 section 7.1.3

How Quantum Programming in Depth covers it

How deep it goes
Derives arbitrary-eigenphase kickback, measurement decoding, repeated-run estimation, binary-digit extraction, and multi-control inverse-QFT readout.
Examples
Uses controlled X versus controlled −X, Z eigenstates |0⟩ and |1⟩, INC eigenphases, and marked states 001 and 110.
What it assumes
Assumes familiarity with qubit superposition and controlled gates while briefly motivating eigenstates and relative versus global phase.
What you can do after
Build it from scratch as phase-estimation circuits or as a universal wrapper converting a marking oracle into a phase oracle.
How it connects
Connects controlled unitaries, eigenvectors, eigenvalues, relative phase, phase estimation, phase oracles, and quantum Fourier transform.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Quantum Programming in Depth. The section links point to the parts of the book it draws on.