I am often asked what is the best algorithm to solve SVM, to solve Lasso Regression, to solve Logistic Regression, etc. At the same time, a growing number of first-order methods have been recently proposed, making even experts hard to track down the state-of-the-arts.

In this guest post on Princeton University’s OptiML blog, I finally have a chance to answer all these questions properly and simultaneously.

http://www.minimizingregret.com/2016/06/how-to-solve-classification-and-regression.html