Quantum circuit for training perceptron models

Qu&Co comments on this publication:

The perceptron algorithm dates back to the late 1950s and is an algorithm for supervised learning of binary classifiers. In a 2016 paper, Wiebe et al. proposed a quantum algorithm (based on Grover’s quantum-search approach), which can quadratically speed-up the training of a perceptron. In this paper, Zheng et al. describe their design for a quantum-circuit to implement the training-algorithm of Wiebe et al. They also analyze the resource requirements (qubits and gates) and demonstrate the feasibility of their quantum-circuit by testing it on the ibmqx5 (a 16 qubit universal gate quantum processor developed by IBM)