A new quantum algorithm for data analysis has been demonstrated by a team of researchers.
New quantum algorithm is a perfect "fit." Photo credit: Physical Review Letters
Nathan Wiebe, a postdoctoral fellow at the Institute for Quantum Computing (IQC) at the University of Waterloo, has partnered with Daniel Braun of the Université de Toulouse, and Seth Lloyd of MIT, in the collaborative research program. They have described a quantum algorithm that is applicable for a prominent data analysis technique.
Using a quantum computer, the team demonstrated an algorithm that can improve "least-squares fitting." In order to solve linear systems of equations, previous studies have suggested an algorithm. Based on this algorithm, the present research team has developed and suggested an algorithm that can estimate the data-fit quality efficiently. This does not require a complete solution initially. It also does not require complete characterization of the quantum computer’s state.
The algorithm will facilitate quick searches of huge quantum data sets as well as precise and simple approximations.
Wiebe informed that even normal computational problems that can be addressed by classical algorithms can be performed using quantum computers at comparatively faster speed. The study shows an alternative method of studying the quantum state that occurs due to the result of quantum computations.
Methods for certifying quantum computer output or quantum simulator output have been a difficult proposition in quantum information research. The current study will enable better methods. It also suggests that in order to construct huge quantum information processors quantum computers may be necessary.
The study has been published in the journal Physical Review Letters. The American Physical Society has highlighted the study in its website.