arXiv (Cornell University) · 2017 · 103 citations · 0 references
ComputingComputer ScienceData ScienceClinical Decision SupportMachine LearningPredictive AnalyticsDecision TreesDeep LearningDisease ModelingNeural NetworksProbabilistic Machine LearningRepresentation LearningAccuracyInterpretability Of TreesMachine Learning Prediction ModelsModel InterpretabilityModel RegularizationPredictive Modeling Of SepsisRegularizationSepsis Prediction Models
The lack of interpretability remains a key barrier to the adoption of deep models in many applications. In this work, we explicitly regularize deep models so human users might step through the process behind their predictions in little time. Specifically, we train deep time-series models so their class-probability predictions have high accuracy while being closely modeled by decision trees with few nodes. Using intuitive toy examples as well as medical tasks for treating sepsis and HIV, we demonstrate that this new tree regularization yields models that are easier for humans to simulate than simpler L1 or L2 penalties without sacrificing predictive power.