Intrinsic Quantification of Domain-Shift Magnitude in Reinforcement Learning

Using a meta model to quantify policy suitability for upcoming domain shifts