Realization of a quantum neural network using repeat-until-success circuits in a superconducting quantum processor
W. Vlothuizen (TU Delft - QuTech Advanced Research Centre, TU Delft - BUS/Quantum Delft)
J. F. Marques (Kavli institute of nanoscience Delft, TU Delft - QCD/DiCarlo Lab)
J. van Straten (TU Delft - BUS/Quantum Delft, TU Delft - QuTech Advanced Research Centre)
H. Ali (TU Delft - Building Operations, TU Delft - QuTech Advanced Research Centre, Kavli institute of nanoscience Delft)
N. Muthusubramanian (TU Delft - QN/Kavli Nanolab Delft, TU Delft - QuTech Advanced Research Centre, TU Delft - Communication QuTech, Kavli institute of nanoscience Delft)
C. Zachariadis (TU Delft - QuTech Advanced Research Centre, Kavli institute of nanoscience Delft, TU Delft - QN/Kavli Nanolab Delft)
J. van Someren (TU Delft - QCD/Feld Group, TU Delft - QuTech Advanced Research Centre)
M. Beekman (TU Delft - Codesigning Social Change, TNO, TU Delft - QuTech Advanced Research Centre)
N. Haider (DIANA FEA , Microwave Sensing, Signals & Systems, TU Delft - QuTech Advanced Research Centre, TNO)
A. Bruno (TU Delft - QN/Kavli Nanolab Delft, Kavli institute of nanoscience Delft)
C. G. Almudever (TU Delft - QCD/Sebastiano Lab, TU Delft - QuTech Advanced Research Centre)
L. DiCarlo (Kavli institute of nanoscience Delft, TU Delft - QN/DiCarlo Lab, TU Delft - QuTech Advanced Research Centre, TU Delft - QCD/DiCarlo Lab)
undefined More Authors (External organisation)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Artificial neural networks are becoming an integral part of digital solutions to complex problems. However, employing neural networks on quantum processors faces challenges related to the implementation of non-linear functions using quantum circuits. In this paper, we use repeat-until-success circuits enabled by real-time control-flow feedback to realize quantum neurons with non-linear activation functions. These neurons constitute elementary building blocks that can be arranged in a variety of layouts to carry out deep learning tasks quantum coherently. As an example, we construct a minimal feedforward quantum neural network capable of learning all 2-to-1-bit Boolean functions by optimization of network activation parameters within the supervised-learning paradigm. This model is shown to perform non-linear classification and effectively learns from multiple copies of a single training state consisting of the maximal superposition of all inputs.
help_outline