Principles of Superconducting Quantum Computers

9 Errors and Error Mitigation

9.1 NISQ Processors

As we have mentioned before, the first generation of quantum computers is known as NISQ (Noisy Intermediate-Scale Quantum), a term coined by John Preskill [62]. Intermediate-scale refers to the number of qubits available in fully-programmable systems over the next several years (i.e., the 2020s), in the range of fifty to a few thousand. The number 50 is important, because it is generally considered the point at which classical computers¹ cannot use brute force simulation to replicate the behavior of a quantum system. Therefore, these systems are large enough to do something interesting, potentially demonstrating a computational advantage over classical computing, but are not large enough to employ error correction (see Chapter 10) for long-running, fault-tolerant quantum computing.

In this chapter, we concentrate on the N in NISQ: Noisy. The ability to manipulate matter at a quantum scale is nothing short of remarkable. But the qubits and our ability to control them are not perfect. Theory tells us what will happen in a closed system, with no interaction with the surrounding environment, but this can never be achieved in practice. Noise is a generic term that refers to anything that leads to non-ideal behavior of the system under consideration, including: environmental interactions, imprecise control of qubits, non-ideal measurements, and undesired interactions between qubits. We describe various types of noise that lead to ...