Result Analysis of the NIPS 2003 Feature Selection Challenge

By Isabelle M Guyon, Steve R. Gunn, Asa Ben-Hur, Gideon Dror. 2004

724 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.

In Evaluation & Benchmarks

Citations in the corpus

Scikit-learn: Machine Learning in Python (2011, methodology): “ to maximize computational eﬃciency. In Table 1, we compare computation time for a few algorithms implemented in the major machine learning toolkits accessible in Python. We use the Madelon data set (Guyon et al., 2004), 4400 instances and 500 attributes, The data set is quite large, but small enoug”
Learning Privately over Distributed Features: An ADMM Sharing Approach (2019, methodology): “5 Experiments We test our algorithm by training l2-norm regularized logistic regression on two popular public datasets, namely, a9a from UCI [30] and giette [31].”
Variable selection strategies for nearest neighbor imputation methods used in remote sensing based forest inventory (2012, background): “There are not many studies that address this issue, but there has been some research in machine learning for selecting variables that can be used in classification model (Guyon et al., 2004).”
Analysis of the IJCNN 2007 agnostic learning vs. prior knowledge challenge (2008, background): “The design of the challenge was informed by experience gained from the previous competitions that we have organized [13, 14].”
Moreau-Yosida Regularization for Grouped Tree Structure Learning (2010, methodology): “For each binary classification task, we compute the balanced error rate [3] to cope with the unbalanced positive and negative samples.”
Kernel Methods for Deep Learning (2009, neutral): “helpful, some spurious, and some sparse (Guyon et al., 2005).”
On Symmetric Losses for Learning from Corrupted Labels (2019, methodology): “BER and AUC are popular metrics for imbalanced data classification [Cheng et al., 2002; Guyon et al., 2005].”
SAINT: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pre-Training (2021, neutral): “The Arcene dataset, furnished by UCI [11], comprises anonymized patient records, where the goal is to classify entries as containing cancer patterns or normal patterns [13].”
Boosting Frank-Wolfe by Chasing Gradients (2020, methodology): “We consider the task of recognizing the handwritten digits 4 and 9 from the Gisette dataset (Guyon et al., 2005), available at https://archive.ics.uci.edu/ml/ datasets/Gisette.”
Decoding brain states from fMRI connectivity graphs (2011, background): “Subband 1 sees very little improvement in accuracy, indicating that connectivity in this frequency subband is probably not discriminative between rest and movie conditions.”
A Distributed Frank-Wolfe Algorithm for Communication-Efficient Sparse Learning (2014, methodology): “LASSO regression with distributed features We also conduct experiments on Dorothea, a drug discovery dataset from the NIPS 2003 Feature Selection Challenge (Guyon et al., 2004).”
Feature Importance in Nonlinear Embeddings (FINE): Applications in Digital Pathology (2016, methodology): “The NIPS 2003 Feature Selection Challenge included five datasets, all involving binary classification problems; we used the four datasets that contain non-binary features.”
Feature Selection with the Boruta Package (2010, neutral): “(Guyon et al. 2005) The data set contains 2000 objects corresponding to points located in 32 vertices of a 5-dimensional hypercube.”