Neural Information Processing Systems · 2014 · 100 citations · 11 references
ComputingComputer ScienceData ScienceAlgorithmsMachine LearningStatistical LearningLinear ProgrammingOptimization For Machine LearningOutlier DetectionOutlier HandlingRobust Statistical ModelingRobust StatisticsAdversarial RobustnessBinary ClassificationLogistic RegressionRobust EstimationRobust Regression
We consider logistic regression with arbitrary outliers in the covariate matrix. We propose a new robust logistic regression algorithm, called RoLR, that estimates the parameter through a simple linear programming procedure. We prove that RoLR is robust to a constant fraction of adversarial outliers. To the best of our knowledge, this is the first result on estimating logistic regression model when the covariate matrix is corrupted with any performance guarantees. Besides regression, we apply RoLR to solving binary classification problems where a fraction of training samples are corrupted.
11
Residuals and Influence in Regression
Robert F. Ling, R. Dennis Cook, Sanford Weisberg · Technometrics · 1984 · 2.7K citations · Full text
Logistic Regression Diagnostics
Daryl Pregibon · The Annals of Statistics · 1981 · 1.3K citations · Full text
EfficientL 1 regularized logistic regression
Sun-In Lee, Honglak Lee, Pieter Abbeel et al. · 2006 · 287 citations