SurvITE: Learning Heterogeneous Treatment Effects from Time-to-Event Data

Neural Information Processing Systems 

We study the problem of inferring heterogeneous treatment effects from time-to-event data.