Finding Low-Rank Solutions via Nonconvex Matrix Factorization, Efficiently and Provably
Finding Low-Rank Solutions via Nonconvex Matrix Factorization, Efficiently and Provably is a scholarly work, published in 2018 in ''SIAM journal on imaging sciences''. The main subjects of the publication include sublinear function, mathematics, gradient descent, rank, matrix, matrix completion, factorization, convex function, mathematical optimization, initialization, biological function, regular polygon, discrete mathematics, compressed sensing, combinatorics, and machine learning. The authors propose the Bi-Factored Gradient Descent (BFGD) algorithm, an efficient first-order method that operates on the $U, V$ factors.