Variational Inference for Sparse and Undirected Models

35 citations recorded by the source corpus.

How researchers see this paper

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

Efficient and Scalable Bayesian Neural Nets with Rank-1 Factors (2020, background): “This removes posterior correlations between the weights which can be otherwise difﬁcult to approximate (Ingraham & Marks, 2017; Louizos et al., 2017).”
Bayesian Compression for Deep Learning (2017, methodology): “It should be mentioned that a similar form of non-centered variational inference for the horseshoe has been also successfully employed for undirected models at [Ingraham and Marks, 2016].”
Variational auto-encoding of protein sequences (2017, methodology): “Additionally, variational inference on graphical models (akin to those presented above as Potts models) were shown to hold promise in predicting protein structure [7].”
Structured Variational Learning of Bayesian Neural Networks with Horseshoe Priors (2018, methodology): “Adopting non-centered parameterizations (Ingraham & Marks, 2016), helps alleviate the issue.”
Model Selection in Bayesian Neural Networks via Horseshoe Priors (2017, background): “While their favorable sparsity inducing properties stem from this coupling, it also gives rise to coupled posteriors that exhibit pathological funnel shaped geometries (Betancourt and Girolami, 2015; Ingraham and Marks, 2017) that are difficult to reliably sample or approximate.”