Direct Fidelity Estimation

Direct Fidelity Estimation

Direct Fidelity Estimation (DFE) is an efficient method for verifying how close the state actually produced by a quantum device is to the intended target state, without the exponential cost of full quantum state tomography. The fidelity between a desired pure state and the actual (possibly noisy) state is a unitless number between 0 and 1 that quantifies their overlap, and measuring it directly is essential for benchmarking and debugging quantum hardware. Rather than reconstructing the entire density matrix, DFE estimates the fidelity from a small number of randomly chosen Pauli measurements, sampled according to their importance, so the number of measurements scales far more favorably with system size. This Cirq example implements the DFE protocol for a given circuit: it computes the Pauli expectation values that characterize the target state, samples the most informative Pauli operators, measures them on the (optionally noisy) device or simulator, and combines the results into an unbiased fidelity estimate with quantified statistical error. It is a practical, resource-efficient characterization tool and a worked example of how randomized measurement schemes make device verification tractable at scale.

Run it

pip install -r requirements.txt
python direct_fidelity_estimation.py

Source and license

Imported from examples/direct_fidelity_estimation.py in quantumlib/Cirq at v1.6.1, under the Apache License 2.0. Original authors: The Cirq Developers. The upstream LICENSE is included alongside this example.