1: \begin{abstract}

2: Consider the problem of minimizing the expected value of a cost function parameterized by a random variable. The classical sample average approximation (SAA) method for solving this problem requires minimization of an ensemble average of the objective at each step, which can be expensive. % One of the major drawbacks of the SAA method is that each step of the algorithm involves solving an optimization problem which might be computationally expensive.

3: In this paper, we  propose a stochastic successive upper-bound minimization method (SSUM) which minimizes an \emph{approximate} ensemble average at each iteration. To ensure convergence and to facilitate computation, we require the approximate ensemble average to be a locally tight upper-bound of the expected cost function and be easily optimized. The main contributions of this work include the development and analysis of the SSUM method as well as its applications in linear transceiver design for wireless communication networks and online dictionary learning. Moreover, using the SSUM framework, we extend the classical stochastic (sub-)gradient (SG) method to the case of minimizing a nonsmooth nonconvex objective function and establish its convergence. %it is shown that both the SAA method and the stochastic gradient (SG) method are special cases of the SSUM framework. %For the stochastic wireless beamforming problem, our numerical experiments indicate that the SSUM approach significantly outperforms the other existing algorithms such as the SA method in terms of the achievable ergodic sum rate in the network.

4: 

5: 

6: \keywords{Stochastic Successive Upper-bound Minimization \and  Stochastic Successive Inner Approximation \and Sample Average Approximation \and Stochastic Beamformer Design}

7: \end{abstract}

8: