Prediction Models That Learn to Avoid Missing Values
This provides practitioners with a tool to maintain interpretability when dealing with test-time missing values, though it represents an incremental improvement over existing methods.
The paper tackles the challenge of handling missing values at test time while maintaining both accuracy and interpretability by proposing missingness-avoiding (MA) machine learning, a framework that trains models to rarely require missing feature values during testing. Experiments on real-world datasets show that MA variants of decision trees, LASSO, random forests, and gradient boosted trees reduce reliance on missing features while maintaining competitive predictive performance.
Handling missing values at test time is challenging for machine learning models, especially when aiming for both high accuracy and interpretability. Established approaches often add bias through imputation or excessive model complexity via missingness indicators. Moreover, either method can obscure interpretability, making it harder to understand how the model utilizes the observed variables in predictions. We propose missingness-avoiding (MA) machine learning, a general framework for training models to rarely require the values of missing (or imputed) features at test time. We create tailored MA learning algorithms for decision trees, tree ensembles, and sparse linear models by incorporating classifier-specific regularization terms in their learning objectives. The tree-based models leverage contextual missingness by reducing reliance on missing values based on the observed context. Experiments on real-world datasets demonstrate that MA-DT, MA-LASSO, MA-RF, and MA-GBT effectively reduce the reliance on features with missing values while maintaining predictive performance competitive with their unregularized counterparts. This shows that our framework gives practitioners a powerful tool to maintain interpretability in predictions with test-time missing values.