Optimised designs for fault-tolerant quantum computers
Lead Research Organisation:
University of Sheffield
Department Name: Physics and Astronomy
Abstract
Quantum systems can store and process quantum information, opening up the prospect of new technologies that outperform conventional supercomputers in many areas. However, quantum information is more fragile than classical binary information, being more susceptible to noise and rapid degradation. To build a reliable device, quantum information must be stored within an abstract quantum codespace that protects it against noise. Quantum computers must also tolerate faults occurring while processing information. This project will develop new techniques for fault-tolerantly storing and processing quantum information, including development and implementation of a new compiler for compressing quantum circuits into the fewest possible number of gates.
Organisations
People |
ORCID iD |
| Earl Campbell (Primary Supervisor) | |
| Luke Heyfron (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509735/1 | 01/10/2016 | 30/09/2021 | |||
| 1798034 | Studentship | EP/N509735/1 | 01/10/2016 | 27/02/2021 | Luke Heyfron |
| Description | A novel quantum compiling algorithm has been developed that optimizes an important metric known as the "T count" that approximates the cost of executing quantum circuits on fault tolerant quantum architectures. The effectiveness of the algorithm was verified by first coding a quantum compiler framework in C++ (called TOpt) where the T count optimization step is an interchangeable subroutine. The T counts were compared for various algorithms using benchmark circuits that perform useful functions as well as random circuits. In most cases, the novel T count optimization algorithm, which we call TODD, outperformed all previous algorithms in terms of raw T count. So the TODD algorithm, along with its output quantum circuits, are key findings associated with this award. A similar algorithm was developed for qudit-based quantum computers (i.e. a quantum computer which uses d-level quantum systems rather than 2-level qubits) and it was used to obtain a unitary circuit for the Toffoli gate (a doubly-controlled inverter) with a T count of 4 where the best previous had a T count of 7. |
| Exploitation Route | The TOpt compiler is freely available as an open source command line application from the above GitHub address. This allows physicists, computer scientists and other interested parties to use it to optimize quantum circuits and take the source and continue developing the tools according to their research needs. While the T count optimization algorithms are effective, there are avenues for further work on improving technical aspects of the algorithms as outlined in the above paper and pre-print article. This is especially true for the harder problem of qudit optimization. |
| Sectors | Digital/Communication/Information Technologies (including Software) |
| URL | https://doi.org/10.1088/2058-9565/aad604,https://github.com/Luke-Heyfron/TOpt,https://arxiv.org/abs/1902.05634 |