TY - JOUR
T1 - Efficient Discrete Feature Encoding for Variational Quantum Classifier
AU - Yano, Hiroshi
AU - Suzuki, Yudai
AU - Itoh, Kohei M.
AU - Raymond, Rudy
AU - Yamamoto, Naoki
N1 - Funding Information:
This work was supported by MEXT Quantum Leap Flagship Program under Grant JPMXS0118067285 and Grant JPMXS0120319794.
Publisher Copyright:
© 2021 by the Author(s).
PY - 2021
Y1 - 2021
N2 - Recent days have witnessed significant interests in applying quantum-enhanced techniques for solving a variety of machine learning tasks. Variational methods that use quantum resources of imperfect quantum devices with the help of classical computing techniques are popular for supervised learning. Variational quantum classification (VQC) is one of such methods with possible quantum advantage in using quantum-enhanced features that are hard to compute by classical methods. Its performance depends on the mapping of classical features into a quantum-enhanced feature space. Although there have been many quantum-mapping functions proposed so far, there is little discussion on efficient mapping of discrete features, such as age group, zip code, and others, which are often significant for classifying datasets of interest. We first introduce the use of quantum random-access coding (QRAC) to map such discrete features efficiently into limited number of qubits for VQC. In numerical simulations, we present a range of encoding strategies and demonstrate their limitations and capabilities. We experimentally show that QRAC can help speeding up the training of VQC by reducing its parameters via saving on the number of qubits for the mapping. We confirm the effectiveness of the QRAC in VQC by experimenting on classification of real-world datasets with both simulators and real quantum devices.
AB - Recent days have witnessed significant interests in applying quantum-enhanced techniques for solving a variety of machine learning tasks. Variational methods that use quantum resources of imperfect quantum devices with the help of classical computing techniques are popular for supervised learning. Variational quantum classification (VQC) is one of such methods with possible quantum advantage in using quantum-enhanced features that are hard to compute by classical methods. Its performance depends on the mapping of classical features into a quantum-enhanced feature space. Although there have been many quantum-mapping functions proposed so far, there is little discussion on efficient mapping of discrete features, such as age group, zip code, and others, which are often significant for classifying datasets of interest. We first introduce the use of quantum random-access coding (QRAC) to map such discrete features efficiently into limited number of qubits for VQC. In numerical simulations, we present a range of encoding strategies and demonstrate their limitations and capabilities. We experimentally show that QRAC can help speeding up the training of VQC by reducing its parameters via saving on the number of qubits for the mapping. We confirm the effectiveness of the QRAC in VQC by experimenting on classification of real-world datasets with both simulators and real quantum devices.
KW - Discrete features
KW - quantum machine learning
KW - quantum random-access coding (QRAC)
KW - supervised learning
KW - variational quantum algorithms
UR - http://www.scopus.com/inward/record.url?scp=85130143462&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85130143462&partnerID=8YFLogxK
U2 - 10.1109/TQE.2021.3103050
DO - 10.1109/TQE.2021.3103050
M3 - Article
AN - SCOPUS:85130143462
SN - 2689-1808
VL - 2
JO - IEEE Transactions on Quantum Engineering
JF - IEEE Transactions on Quantum Engineering
M1 - 3103214
ER -