Active Learning for DecisionMaking from Imbalanced Observational Data
[edit]
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:60466055, 2019.
Abstract
Machine learning can help personalized decision support by learning models to predict individual treatment effects (ITE). This work studies the reliability of predictionbased decisionmaking in a task of deciding which action $a$ to take for a target unit after observing its covariates $\tilde{x}$ and predicted outcomes $\hat{p}(\tilde{y} \mid \tilde{x}, a)$. An example case is personalized medicine and the decision of which treatment to give to a patient. A common problem when learning these models from observational data is imbalance, that is, difference in treated/control covariate distributions, which is known to increase the upper bound of the expected ITE estimation error. We propose to assess the decisionmaking reliability by estimating the ITE model’s Type S error rate, which is the probability of the model inferring the sign of the treatment effect wrong. Furthermore, we use the estimated reliability as a criterion for active learning, in order to collect new (possibly expensive) observations, instead of making a forced choice based on unreliable predictions. We demonstrate the effectiveness of this decisionmaking aware active learning in two decisionmaking tasks: in simulated data with binary outcomes and in a medical dataset with synthetic and continuous treatment outcomes.
Related Material


