Abstract
News editors obtained the following quote from the background information suppli ed by the inventors:“A machine learning model may not perform as well during de ployment in production as it did whenvalidating it offline. This can be because the distribution of data that a machine learning model is exposedto can change over time, commonly referred to as data drift or feature drift. The drift can b e gradual orcan happen overnight and cause the model’s performance or output to change. Tracking changes in dataand understanding how these changes affect mod el performance is critical to understanding if a machinelearning model is worki ng appropriately.”
NETL
NSTL