Hybrid Quantum-Classical Neural Network Transfer Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high learning cost and computational expense associated with achieving highly accurate results using the variational quantum eigensolver (VQE) for quantum chemistry applications, such as drug discovery and material development, due to the need for frequent measurement sampling in noisy intermediate-scale quantum (NISQ) devices.
Innovation Solution
A quantum circuitry learning method that employs a hybrid quantum-classical neural network (HQCNN) with a measurement layer interposed between parameterized quantum circuitry layers, allowing for transfer learning and reducing the number of parameters to be optimized, thereby lowering the learning cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If VQE is used to obtain highly accurate results in quantum chemistry computing, then measurement precision is improved, but learning cost increases due to the need for frequent measurement sampling in NISQ devices
Solution Approach 1:
The patent applies preliminary action by performing transfer learning from a source task to a target task. The quantum circuit parameters optimized for the source task are transferred and reused as initial parameters for the target task, eliminating the need to perform complete optimization from scratch. This preliminary utilization of pre-trained parameters significantly reduces the learning cost and measurement sampling requirements while maintaining high accuracy in quantum chemistry computing.
2Measurement precision
If the number of parameters in quantum circuitry is increased to improve accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the quantum learning problem into two distinct parts: a source task and a target task. By dividing the overall problem, the system can optimize parameters for the source task once and then transfer these optimized parameters to the target task. This segmentation allows the target task to achieve high accuracy without requiring a large number of parameters, as the transferred parameters provide a strong initial foundation that requires minimal further optimization.
Data Source
AI summary
A quantum circuitry learning method comprising: reading an optimized parameter θ* assigned to a parameterized quantum circuitry U(θ*) of a trained first HQCNN, the first HQCNN being trained based on a first data set regarding classical data b1; transferring the parameter θ* to a second feature extraction circuitry F (b2, θ*) included in a second HQCNN; training the second HQCNN based on a second data set regarding the classical data b2 while fixing the parameter θ*, and optimizing a second parameter Φ of a parameterized quantum circuitry U(Φ) of the second HQCNN.