Faulty quantum computation can result in reliable classical outputs
Gerald N. Gilbert, Michael Hamrick, Yaakov S. Weinstein · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2007
The model of quantum computation developed by Kitaev (1 ,∮4.1) shows that a perfect, error-free, quantum computer can lead to reliable classical outputs, despite the need to apply a necessarily probabilistic measurement. In this paper we extend the analysis to account for necessarily imperfect quantum computation. The analysis presented here is required to establish the utility of practical quantum computation even given the assumption that fault-tolerance techniques are successfully applied. This is due to the fact that the application of currently known fault-tolerance techniques does not permanently and completely remove errors. To this end we have introduced a mathematical relation that compares the accuracy of a necessarily imperfect quantum computation to a prescribed performance bound. Finally, we discuss several mathematical aspects of this bound and its usefulness in analyzing quantum computing protocols.