Abstract

We consider a class of stochastic nonlinear programs for which an approximation to a locally optimal solution is specified in terms of a fractional reduction of the initial cost error. We show that such an approximate solution can be found by approximately solving a sequence of sample average approximations. The key issue in this approach is the determination of the required sequence of sample average approximations as well as the number of iterations to be carried out on each sample average approximation in this sequence. We show that one can express this requirement as an idealized optimization problem whose cost function is the computing work required to obtain the required error reduction. The specification of this idealized optimization problem requires the exact knowledge of a few problems and algorithm parameters. Since the exact values of these parameters are not known, we use estimates, which can be updated as the computation progresses. We illustrate our approach using two numerical examples from structural engineering design.

6. Conclusions

We have demonstrated that solving an auxiliary, efficient diagonalization problem to obtain a diagonalization strategy can reduce the overall computing times in stochastic nonlinear programming. In particular, this approach eliminates the need for determining algorithm parameters by means of guesswork or costly numerical experimentation. Instead, the efficient diagonalization problem determines sample sizes and numbers of iterations at each stage using estimated values of cost-to-go, rate of convergence, and sampling error. Even using Matlab, the solution of the diagonalization problem requires only seconds of computing time. Our computational experience indicates that the advantage of an efficient diagonalization approach is more substantial for larger, more complicated problems and when a high-precision solution is sought.