Demonstration of a Quantum Circuit Design Methodology for Multiple Regression
Sanchayan Dutta, Adrien Suau, Sagnik Dutta, Suvadeep Roy, Bikash K. Behera, Prasanta Kumar Panigrahi · arXiv (Cornell University) · 2018
Multiple linear regression, one of the most fundamental supervised learning algorithms, assumes an imperative role in the field of machine learning. In 2009, Harrow et al. [Phys. Rev. Lett. 103, 150502 (2009)] showed that their algorithm could be used to sample the solution of a linear system $\mathbf{Ax=b}$ exponentially faster than any existing classical algorithm. Remarkably, any multiple linear regression problem can be reduced to a linear system of equations problem. However, finding a practical and efficient quantum circuit for the quantum algorithm in terms of elementary gate operations is still an open topic. Here we put forward a 7-qubit quantum circuit design, based on an earlier work by Cao et al. [Mol. Phys. 110, 1675 (2012)], to solve a 3-variable regression problem, utilizing only basic quantum gates. Furthermore, we discuss the results of the Qiskit simulation for the circuit and explore certain possible generalizations to the circuit.