This recovery setting is central to the theoretical understanding of dictionary learning. We give the first efficient algorithm that provably recovers mb A_0 ...

Apr 26, 2015 · Our algorithmic pipeline centers around solving a certain nonconvex optimization problem with a spherical constraint, and hence is naturally ...

Jun 17, 2016 · Theorem. For any θ ∈ (0,1/3), given Y = A0X0 with A0 a complete dictionary and X0 ∼i.i.d. BG(θ), there is a polynomial time.

This work gives the first efficient algorithm that provably recovers A0 when X0 has O(n) nonzeros per column, under suitable probability model for X0, ...

Nov 11, 2015 · Our algorithmic pipeline centers around solving a certain nonconvex optimization problem with a spherical constraint. In this paper, we ...

Dive into the research topics of 'Complete dictionary recovery using nonconvex optimization'. Together they form a unique fingerprint. Sort by; Weight ...

Nov 23, 2016 · Our algorithmic pipeline centers around solving a certain nonconvex optimization problem with a spherical constraint, and hence is naturally ...

We consider the problem of recovering a complete (i.e., square and invertible) dictionary mbA0, from mbY=mbA0mbX0 with mbY∈Rn×p. This recovery setting is ...

This algorithm is based on nonconvex optimization. Our proofs give a geometric characterization of the high-dimensional objective landscape, which shows ...

Our algorithm is based on nonconvex optimization with a spherical constraint, and hence is naturally phrased in the language of manifold optimization. Our ...