Lauren A. Hannah

Evolutionary Policy Iteration Under a Sampling Regime for Stochastic Combinatorial Optimization

This article modifies the evolutionary policy selection algorithm of Chang et al., which was designed for use in infinite horizon Markov decision processes (MDPs) with a large action space to a discrete stochastic optimization problem, in an algorithm called Evolutionary Policy Iteration-Monte Carlo (EPI-MC). EPI-MC allows EPI to be used in a stochastic combinatorial optimization setting with a finite action space and a noisy cost (value) function by introducing a sampling schedule. Convergence of EPI-MC to the optimal action is proven and experimental results are given.

SEMICONVEX REGRESSION FOR METAMODELING-BASED OPTIMIZATION ∗

Stochastic search involves finding a set of controllable parameters that minimizes an unknown objective function using a set of noisy observations. We consider the case when the unknown function is convex and a metamodel is used as a surrogate objective function. Often the data are non-i.i.d. and include an observable state variable, such as applicant information in a loan rate decision problem. State information is difficult to incorporate into convex models. We propose a new semiconvex regression method that is used to produce a convex metamodel in the presence of a state variable. We show consistency for this method. We demonstrate its effectiveness for metamodeling on a set of synthetic inventory management problems and a large real-life auto loan dataset.