Back
Qubit-aware quantum algorithm development for industrial use cases
The overall goal of our work is to identify areas of industrial applications for quantum algorithms, for which the quantum resources needed are minimal while still witnessing quantum speed-up compared to state-of-the-art classical methods. Importantly, we will not just try out another implementation of a quantum algorithm for which there is no analytical evidence that it will scale with a significant quantum advantage. Rather, the idea is to be aware of industrially relevant state-of-the-art classical methods and derive long-lasting academic results that inform for what problems of practical interest one really wants to eventually run quantum algorithms. To achieve that, we propose to use classical hardware and algorithm whenever possible and to solely employ quantum computers for highly specialized sub-routines when there are indications for potential large quantum speed-up. That is, we will further develop hybrid classicalquantum architectures that shoot for at least super-quadratic speed-up—in order for mid-term quantum technologies to have any hope to witness quantum advantage [3]. We then propose to work in an early fault-tolerant model where reasonably high fidelity logical qubits and quantum gates are available [6], but qubit count is the all dominating and most expensive metric. We aim to develop flexible qubit aware schemes that can trade qubit count for parallelizable sample complexity—as well as algorithm depth and other resources that can come at variable cost. Whereas some work in this direction has started to emerge [15], it is in our opinion a highly understudied subject, with the potential for strong industrial impact.
Field of action:
Quantum Computing
- Faculty 1 – Mathematics, Computer Science and Natural Sciences
Address:
Insitut for Quantum Information, 109 Modulbau Physik 2, Campus Melaten, Otto-Blumenthal-Straße 20, 52074 Aachen, German
Contact:
Prof. Dr. Mario Berta
berta@physik.rwth-aachen.de
Status:
in preparation