Complex Rotation Quantum Dynamic Neural Networks (CRQDNN) using Complex Quantum Neuron (CQN): Applications to time...

Area of Science:

• Quantum Computing
• Artificial Intelligence
• Neural Networks

Background:

• Quantum Neural Networks (QNN) leverage quantum entanglement for novel computation.
• Existing QNN models often underutilize quantum entanglement due to reliance on real quantum operations.

Purpose of the Study:

• Propose a Complex Quantum Neuron (CQN) model for deep quantum entanglement.
• Introduce a hybrid Complex Rotation Quantum Dynamic Neural Network (CRQDNN) integrating CQN and classical neurons.
• Develop a model for effective time series prediction using enhanced quantum entanglement.

Main Methods:

• The Complex Quantum Neuron (CQN) model enables deep quantum entanglement.
• The Complex Rotation Quantum Dynamic Neural Network (CRQDNN) is a three-layer hybrid architecture.
• An infinite impulse response (IIR) filter provides memory for time series processing.
• The Levenberg-Marquardt (LM) algorithm facilitates rapid parameter learning.

Main Results:

• The CRQDNN model demonstrates capability in processing time series data.
• Successful application in chaotic time series prediction.
• Effective application in predicting electronic remaining useful life (RUL).

Conclusions:

• The proposed CQN and CRQDNN models offer a novel approach to quantum neural networks.
• These models enhance the exploitation of quantum entanglement for computational tasks.
• The developed network is effective for diverse time series prediction applications.