Rapid Learning with Stochastic Focus of Attention

Raphael Pelossof, Zhiliang Ying

We present a method to stop the evaluation of a decision making process when the result of the full evaluation is obvious. This trait is highly desirable for online margin-based ma…