Nonnegative/binary matrix factorization with a D-Wave quantum annealer
D. O'Malley, V.V. Vesselinov, B. Alexandrov, L. Alexandrov
DOE LANL Presentation2017
Summary
Matrix factorization is a powerful unsupervised machine learning technique that can be used to extract hidden features from complex datasets. In this presentation, we demonstrate the use of a D-Wave quantum annealer to perform nonnegative/binary matrix factorization, which is a specific type of matrix factorization that imposes nonnegativity and binary constraints on the factors. We show how this approach can be applied to analyze complex datasets and extract meaningful features, with examples from geothermal, geochemical, and climate applications.