In compressed sensing, a measurement matrix Φ having low coherence with sparse
dictionary Ψ can achieve better signal reconstruction performance. To improve the signal …

Deep learning architectures employ various regularization terms to handle different types of
priors. Non-smooth regularization terms have shown promising performance in the deep …

We present a unitary approach to the design of incoherent frames and to dictionary learning,
by using a single function that promotes incoherence for both problems. This function has a …

In compressed sensing (CS), coherence is a simple and practical measure of the quality of a
sensing matrix. The smaller the coherence of the sensing matrix, the better the …

Structuring the latent space in probabilistic deep generative models, eg, variational
autoencoders (VAEs), is important to yield more expressive models and interpretable …