The Quantum Architecture, Software, and Informatics (or quasi) group is an informal division of the Foundations of Software Systems group, in the Department of Informatics at the University of Sussex.
We investigate high-level architectures for scalable quantum computation, develops software platforms to do practical quantum computation, back by research into the theory of quantum information and quantum computation.
We are a part of the Sussex Centre for Quantum Technologies.
What resources are important in a quantum computation, and what techniques can we use to reduce the use of those resources?
What challenges are there when attempting to build a scalable quantum computing system, and how can we effectively address them?
How can we verify the correctness of a quantum computation, and how difficult is it to simulate quantum computations on more conventional computers?
What are the appropriate mathematical tools to investigate questions in quantum computation, and how can we develop them further?
How can we practically and effectively describe quantum computations? We are working together with the QISKit team to develop OpenQASM 3 — a revision of IBM's successful OpenQASM 2 circuit description language. OpenQASM 3 will allow the programmer to work more closely with quantum hardware platforms, but also will provide ways for programmers to write more versatile code to describe more general quantum computations.
In some quantum architectures, the physical systems which used to store qubits may not be mobile. This represents an obstacle to realising two-qubit operations on pairs of qubits which are not immediately next to one another. Entanglement swapping is one technique to allow 'remote' qubits to interact, using a chain of qubits as intermediaries. However, as scalable quantum computation will rely on being able to perform operations in parallel on different qubits, problems may arise if we would like to use qubits on more than one overlapping chain. With Steven Herbert, we demonstrate how the theory of linear network coding may be used to help realise simultaneous entanglement swapping between multiple qubit pairs.
Diagrammatic calculi are systems for using diagrams to carry out rigorous calculations. The ZX calculus and the ZH calculus are two such 'calculi', which are useful to represent quantum circuits as tensor networks, and to 'rewrite' them to perform simplications and prove equalities. However, the pre-existing versions of these calculi involve rules which require extra bookkeeping to track scalar factors, particularly if the two systems of diagrams are to be used together. We devised an alternative presentation of these rewrite systems, making it easier to use them either together or individually.
There are certain 'simple' operations which a quantum computer must be able to perform to have any computational advantage over classical computers. These operations may be costly to realise in a fault-tolerant manner. To reduce the cost of a computation, it is useful to try to find techniques to reduce the number of such costly operations. Together with Quanlong Wang and Xiaoning Bian, we described and implemented techniques which set new records for the reduction of the number of 'T gates' in a quantum computation. To do so, we used the ZX calculus to describe this problem in terms of a simple optimisation problem, and set out a framework of approaches to find approximate solutions to this problem based on circuit identities.
Dr. de Beaudrap is a Lecturer in Theoretical Computer Science in the Department of Informatics, and a member of the Foundations of Software Systems group. His research interests range from theoretical physics to pure mathematics (number theory, algebra, and cominatorics) and computational complexity (efficient algorithms and counting complexity).
At the moment, quasi is a new group, and many of the uses of the words 'we' and 'our' here should be understood to be the 'royal we'.