Grafting: Fast, Incremental Feature Selection by Gradient Descent in Function Space

By Simon Perkins, Kevin Lacker, James Theiler. 2003

384 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

A Survey on Malware Detection Using Data Mining Techniques (2017, methodology): “…some embedded methods perform more efﬁciently than wrapper methods by directly optimizing an objective function, often deﬁned by two or more parameters – one to encourage the goodness-of-ﬁt and the other to penalize for a large number of variables [Rakotomamonjy 2003; Perkins et al. 2003].”
Hinge-Loss Markov Random Fields and Probabilistic Soft Logic (2015, methodology): “Other approaches to structure learning search directly over grounded PGMs, including `1-regularized pseudolikelihood maximization (Ravikumar et al., 2010b) and grafting (Perkins et al., 2003; Zhu et al., 2010).”
Graph Regularized Sparse Coding for Image Representation (2011, methodology): “In the following, we introduce an optimization method based upon coordinate descent to solve this problem.”
An introduction to variable and feature selection (2003, methodology): “The sensitivity of the objective function to changes in wi (method 3) is used to devise a forward selection procedure in one paper presented in this issue (Perkins et al., 2003).”
A Comparison of Optimization Methods and Software for Large-scale L1-regularized Linear Classification (2010, methodology): “Perkins et al. (2003) proposed an active set method for (1) with logistic loss.”
Gradient boosted feature selection (2014, unknown): “Learning”
Histopathological Image Analysis: A Review (2009, background): “More recent feature selection research has focused on such methods as genetic algorithms, simulated annealing, boosting [99] and grafting [100].”
Applying Conditional Random Fields to Japanese Morphological Analysis (2004, background): “(Perkins et al., 2003) reported that L1-regularizer should be chosen for a problem where most of given features are irrelevant .”
Automatic Feature Learning for Spatio-Spectral Image Classification With Sparse SVM (2014, methodology): “to the online feature selection algorithm described in [41].”
Fast Optimization Methods for L1 Regularization: A Comparative Study and Two New Approaches (2007, methodology): “Gauss-Seidel [16] N Y N 5 1 Shooting [15] N Y N 7 1 Grafting [6] N Y N 4 2 Sub-Gradient N Y N 8 2”
Efficient Structure Learning of Markov Networks using L1-Regularization (2006, methodology): “[19]; given correct estimates of t he gradient, this algorithm is guaranteed to terminate only at the unique global optimum, for any reaso n ble feature introduction method.”
Feature Selection (2016, methodology): “Their method is based on a stagewise gradient descent regularized risk framework (Perkins et al. 2003).”
Speech Recognition Using Augmented Conditional Random Fields (2009, background): “general models as the gradient is defined and calculated [51].”