We propose a randomized gradient method - or a randomized cutting-plane method from a dual viewpoint. From the primal viewpoint, our method bears a resemblance to the stochastic approximation family. But in contrast to stochastic approximation, the present method builds a model problem.
Kecskemet College, Pallasz Athene University. Izsaki ut 10, 6000 Kecskemet, Hungary; and Budapest University of Technology and Economics. Muegyetem rkp 3-9, 1111 Budapest, Hungary.