pith. sign in

arxiv: 1902.03755 · v1 · pith:HHMDH4PFnew · submitted 2019-02-11 · 📊 stat.ML · cs.LG· math.OC

Efficient Primal-Dual Algorithms for Large-Scale Multiclass Classification

classification 📊 stat.ML cs.LGmath.OC
keywords multiclassalgorithmsefficienthingelossesparticularaccuracyalgorithm
0
0 comments X
read the original abstract

We develop efficient algorithms to train $\ell_1$-regularized linear classifiers with large dimensionality $d$ of the feature space, number of classes $k$, and sample size $n$. Our focus is on a special class of losses that includes, in particular, the multiclass hinge and logistic losses. Our approach combines several ideas: (i) passing to the equivalent saddle-point problem with a quasi-bilinear objective; (ii) applying stochastic mirror descent with a proper choice of geometry which guarantees a favorable accuracy bound; (iii) devising non-uniform sampling schemes to approximate the matrix products. In particular, for the multiclass hinge loss we propose a \textit{sublinear} algorithm with iterations performed in $O(d+n+k)$ arithmetic operations.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.