Single-Qubit Quantum Neural Network Achieves Efficient Regression and Classification
A recent paper titled "Regression and Classification with Single-Qubit Quantum Neural Networks" has been published on arXiv, under identifier 2412.09486. This study presents a Single-Qubit Quantum Neural Network (SQQNN) that employs parameterized single-qubit unitary operators and quantum measurements. It introduces an innovative data uploading technique and utilizes gradient descent for regression tasks. For classification, a unique training method inspired by polynomial regression achieves global minimization in a single step. The SQQNN demonstrates near error-free performance across various applications, showcasing its efficiency in both regression and classification tasks.
Key facts
- The paper is titled 'Regression and Classification with Single-Qubit Quantum Neural Networks'.
- It is available on arXiv with identifier 2412.09486.
- The SQQNN uses parameterized single-qubit unitary operators and quantum measurements.
- A new data uploading technique is introduced.
- Gradient descent is used for regression tasks.
- A novel training method inspired by polynomial regression is used for classification.
- The classification method finds a global minimizer in a single step.
- The SQQNN exhibits virtually error-free performance across various applications.
Entities
Institutions
- arXiv