Quantum Characterization, Verification, and Validation
This is an advance summary of a forthcoming article in the Oxford Research Encyclopedia of Physics. Please check back later for the full article.
Quantum systems may outperform current digital technologies at various information processing tasks, such as simulating the dynamics of quantum systems and integer factorization. Quantum Characterization, Verification, and Validation (QCVV) is the procedure for estimating the quality of physical quantum systems for use as information processors. QCVV consists of three components.
Characterization means determining the effect of control operations on a quantum system, and the nature of external noise acting on the quantum system. The first characterization experiments (Rabi, Ramsey, and Hahn-echo) were developed in the context of nuclear magnetic resonance. As other effective two-level systems with varying noise models have been identified and couplings become more complex, additional techniques such as tomography and randomized benchmarking have been developed specifically for quantum information processing.
Verification involves verifying that a control operation implements a desired ideal operation to within a specified precision. Often, these targets are set by the requirements for quantum error correction and fault-tolerant quantum computation in specific architectures.
Validation is demonstrating that a quantum information processor can solve specific problems. For problems whose solution can be efficiently verified (e.g., prime factorization), validation may involve running a corresponding quantum algorithm (e.g., Shor’s algorithm) and analyzing the time taken to produce the correct solution. For problems whose solution cannot be efficiently verified, for example, quantum simulation, developing adequate techniques is an active area of research.
The essential features that make a device useful as a quantum information processor also create difficulties for QCVV, and specialized techniques have been developed to surmount these difficulties. The field is now entering a mature phase where a broad range of techniques can address all three tasks. As quantum information processors continue to scale up and improve, these three tasks look to become increasingly relevant, and many challenges remain.