Abstract

Stochastic search algorithms have recently also gained a lot of attention in operations research, machine learning and policy search of robot motor skills due to their ease of use and their generality. Yet, many stochastic search algorithms require relearning if the task changes slightly to adapt the solution to the new situation or the new context. Therefore we consider the contextual stochastic search setup. Here, we want to find good parameter vectors for multiple related tasks, where each task is described by a continuous context vector. Hence, the objective might change slightly for each parameter vector evaluation. In this research, we investigate the contextual stochastic search algorithms that can learn from multiple tasks simultaneously.