A method for constructing quantum feedforward neural network method based on classic training

Abstract

The invention provides a method for constructing a quantum feedforward neural network based on classical training. The method comprises the following steps: 1, giving a clear definition of quantum neurons; 2, selecting a specific activation function, and then using a quantum circuit for representing a model of quantum neurons; 3, proposing a quantum feedforward neural network model on the basis ofthe quantum neuron model in the step 2; and 4, providing a classic training method, quantitatively analyzing the effectiveness of the classic training method, completing the construction of the quantum feedforward neural network based on classic training. The problem that in the prior art, the clear definition of the quantum neural network is not unified is relieved through the method, and a quantum neural network model does not have the quantum states of input, output and weight at the same time; the realization of the activation function has no specific quantum circuit representation; the quantum neural network model has no ductility; and the quantum neural network lacks theoretical analysis on the effectiveness of the training process.

Description

technical field [0001] The invention relates to the technical fields of quantum computing and neural network, in particular to a method for constructing a quantum feedforward neural network based on classical training. Background technique [0002] The artificial neural network can be traced back to the neuron model proposed by McCulloch-Pitts (M-P) in 1943. R. Rosenblatt increased the training process on the basis of M-P neurons, thus proposing a perceptron model. So far, artificial neural networks not only have a sound theoretical basis, but also have played an important role in practical applications, covering areas such as pattern recognition, classification problems, and multivariate data analysis. [0003] The idea of ​​quantum neural network was first proposed by Kak in 1995. It is a model combining classical artificial neural network and quantum computing. At present, many quantum neural network models have been developed, some are classical neural networks that us...

AI Technical Summary

Problems solved by technology

At present, many quantum neural network models have been developed, some are classical neural networks that use quantum computing to achieve potential acceleration capabilities; some are completely described by actual physical devices; some are quantum perceptron models; quantum neurons in some quantum neural network models The input and output of the network are quantum states, and the network is not considered to use the output of the quantum neuron as the input of the next layer of neurons; some quantum neural network models do not have a training process; some quantum neural network models do not have a specific training process, only abstract mathematical expressions, etc.

Images