Latent Variable Bayesian Models for Promoting Sparsity

By David Paul Wipf, Bhaskar D. Rao, Srikantan S. Nagarajan. IEEE Transactions on Information Theory, 2011

318 citations recorded by the source corpus.

Subfields: Computational Mechanics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/fdaa6683679294254a05d2acfeebf31f7cca0e7f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Signal Processing

Citations in the corpus

Compressing Neural Networks using the Variational Information Bottleneck (2018, unknown): “Moreover, it has been shown in Wipf et al. (2011) that the function deﬁned as f δ ( x ) inf α> 0 x 2 α + log( α + δ ) (49) will be concave and non-decreasing in | x | for all δ ≥ 0, and therefore, ρ ( µ i,j ; γ i , a i,j ) must be concave and non-decreasing with respect to | µ i,j | since γ i /a…”
Sparse Regularization via Convex Analysis (2017, background): “[60], [63], but they are not designed to maintain cost function convexity.”
Sparse Signal Recovery With Temporally Correlated Source Vectors Using Sparse Bayesian Learning (2011, background): “In this section we discuss some of them.”
Multi-Task Learning for Subspace Segmentation (2015, methodology): “from (Wipf et al., 2011), with update rules contained in the supplementary file.”
Iterative Reweighted 1 and 2 Methods for Finding Sparse Solutions (2010, background): “The Achilles heel of standard, separable penalties is that if we want to retain a global minimum similar to that of (1), we require a highly concave penalty on each xi [29].”
Expectation propagation for neural networks with sparsity-promoting priors (2013, methodology): “Recently, it has also been shown that in connection with linear models the ARD solution is exactly equivalent to a MAP estimate of the coefficients obtained using a particular class of nonfactorized coefficient prior distributions which includes models that have desirable advantages over MAP weight ”