On the sample complexity of parameter estimation in logistic regression with normal design

The logistic regression model is one of the most popular data generation model in noisy binary classification problems. In this work, we study the sample complexity of estimating the parameters of the logistic regression model up to a given $\ell_2$ error, in terms of the dimension and the inverse temperature, with standard normal covariates. The inverse temperature controls the signal-to-noise ratio of the data generation process. While both generalization bounds and asymptotic performance of the maximum-likelihood estimator for logistic regression …

binary classification complexity cs.it cs.lg data design error logistic regression math.it math.st normal parameters popular regression sample stat.ml stat.th study terms work