Abstract: Disclosed is a method comprising accessing, by a machine learning system, a set of data records for a plurality of users, the data records representative of physical statistics measured for each of the plurality of users over a time period. At least a subset of the data records comprises patterns of missing data for at least a portion of the time period. The method also comprises generating a set of masked data records by masking a subset of the data records in accordance with a pattern of natural missingness from a data record. The method also comprises generating, by the machine learning system, a set of learned representations from at least the set of masked data records. Finally, the method comprises fine tuning, by the machine learning system, a machine learning model using the set of learned representations, the machine learning model configured to perform a downstream machine learning task.

Abstract: In an aspect, a computer-implemented method for assembling a pool of high-risk subjects for developing an acute illness is disclosed. The method comprises obtaining, from a plurality of subjects, (i) one or more responses to one or more health queries, and (ii) geographic incidence data for the acute illness. The method next comprises predicting, using a machine learning model, a risk of developing the acute illness for the plurality of subjects based on the one or more responses and the geographic incidence data. The method next comprises identifying the pool of high-risk subjects from the plurality of subjects, wherein the risk of developing the acute illness for each subject of the pool of high-risk subjects satisfies a threshold. Finally, the method comprises outputting the pool of high-risk subjects.