In this paper, we investigate the problem of monitoring patients in an assisted living environment. We apply machine learning techniques for localization and tracking. We consider an environment such as a smart home, assisted living facility, or recovery unit that is equipped with tiny wireless devices which interact with a device carried by the patient. These indoor, multi-room environments are well suited to learning approaches as barriers usually inhibit the operation of systems which calculate location via ranging and multilateration. The location information can be logged over time to monitor a patient?s activity. Based on data collected in experiments using real-life test beds, we conduct simulations comparing the location estimation accuracy of several learning algorithms.