We show that using a remapped version of the decision rule for a standard dynamic stochastic optimization problem, where the mapping is defined by asymptotic limits of the rule as uncertainty goes to zero, produces a solution that is both more robust (in the sense that a stable solution emerges...