Pointwise Tracking the Optimal Regression Function

This paper examines the possibility of a "reject option" in the context of least squares regression. It is shown that using rejection it is theoretically possible to learn "selective" regressors that