SAMOP 2021 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
Q: Fachverband Quantenoptik und Photonik
Q 11: Quantum Information (joint session QI/Q)
Q 11.12: Poster
Mittwoch, 22. September 2021, 16:30–18:30, P
A perceptron quantum gate for quantum machine learning — •Patrick Huber1, Erik Torrontegui2, Johann Haber3, Patrick Barthel1, Juan Jose Garcia Ripoll2, and Christof Wunderlich1,3 — 1Universität Siegen, Walter-Flex-Straße 3, 57068 Siegen — 2Instituto de Física Fundamental IFF-CSIC - Calle Serrano 113b, 28006 Madrid, Spain — 3eleQtron GmbH, Martinshardt 19, 57074 Siegen
As quantum computing advances towards the implementation of noisy intermediate-scale quantum computers (NISQs), the number of applications and scientific use cases keep growing. A recent addition is machine learning. We demonstrate the implementation of a perceptron on an ion-based quantum computer comprised of three qubits, a bias qubit, a control qubit, and a target qubit, the latter of which encodes the output state of the perceptron. The system uses magnetic gradient induced coupling (MAGIC) which allows for the control of the qubits by microwave radiation. The magnetic gradient also induces an Ising-like interaction between individual ions. This property is exploited in order to implement the perceptron. We demonstrate both the working of the basic perceptron quantum gate as predicted in [1], and show that by successive application of the perceptron more sophisticated multi-qubit quantum gates can be implemented easily and straightforwardly.
[1] Unitary quantum perceptron as efficient universal approximator, E. Torrontegui and J. J. García-Ripoll EPL, 125 3 (2019) 30004 DOI: https://doi.org/10.1209/0295-5075/125/30004