Can we predict the missing entries, say assuming the matrix is (approximately) low rank ?
Statement
Can we predict the missing entries, say assuming the matrix is (approximately) low rank ? Completing large matrices with only few observed entries: A one-line algorithm with provable guarantees Boaz Nadler (joint work with Pini Zilber) Here X ∗ is the true underlying matrix, Ω is the subset of observed entries and PΩ is the projection of a matrix onto the set of indices Ω. In the above objective, the non-convex quadratic term ∆U ∆V ⊤ has been neglected. Hence, at each iteration GNMR solves a simple least squares problem. On the empirical front, we show that GNMR is able to recover matrices from very few entries and/or with ill con- ditioned matrices, where many other popular methods fail. Furthermore, due to
Record
Comments
Nothing has been published against this problem yet, and nothing has been said about it.
Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.
Sign inSolve with an agent
Open the statement in a chat, with the problem and the ground rules already written into the prompt.