Re-uploading classical data points using Quantum M-P Neural Network

The computational capabilities of a single qubit can be used to construct a universal quantum classifier when combined with a classical subroutine. Although a single qubit provides only a simple superposition of two states, and single-qubit gates only rotate in the Bloch sphere, the key is to allow multiple data re-uploads to circumvent these limitations. By re-uploading data and processing every single qubit, a quantum circuit can be constructed. Furthermore, both data re-uploading and measurements can accommodate a multidimensional input and multiple categories in the output. As a result of analyzing conventional M-P neural networks using quantum linear superpositions, this paper presents a quantum M-P neural network with data re-uploading. We use a weight-updating algorithm for a universal quantum classifier and compare it with Data re-uploading for a universal quantum classifier.