Regularized Learning with Networks of Features

By Ted Sandler, John Blitzer, Partha P. Talukdar, Lyle Ungar. 2008

53 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.

Citations in the corpus

Regularizing Text Categorization with Clusters of Words (2016, methodology): “We only considered logistic regression because of its interpretation for L2 regularizers as Gaussian prior on the feature weights and following Sandler et al. (2009), we considered a non-diagonal covariance matrix for L2 based on word similarity before moving to group lasso as presented in the…”
Fast PRISM: Branch and Bound Hough Transform for Object Class Detection (2011, background): “Moreover, (17) can be solved efficiently in this primal form (Chapelle 2007; Sandler et al. 2008), which directly yields the desired flat histogram w (instead of w̃).”
Guarding Against Adversarial Domain Shifts with Counterfactual Regularization (2017, unknown): “um over the variances ˙2 i (). The graph for the underlying regularization is formed in the sample space and induced by the identier variable ID, in contrast to graphs formed in feature space as in Sandler et al. (2009), where prior knowledge is used to form the graph by connecting features that sha”
Feature Selection (2016, background): “Since features show certain kinds of dependencies in these cases, we can model them by an undirected graph, where nodes represent features and edges among nodes show the pairwise dependencies between features (Sandler et al. 2009; Kim and Xing 2009; Yang et al. 2012).”
Feature Selection for Classification: A Review (2014, background): “3 Features with Graph Structure We often have knowledge about pair-wise dependencies between features in many realworld applications [58].”