Hypothesis-free discovery from epidemiological data by automatic detection and local inference for tree-based nonlinearities and interactions
This work addresses the need for reliable inference in machine learning for epidemiological discovery, offering a method to improve uncertainty quantification in detecting risk factors, though it appears incremental as it builds on existing techniques like Shapley values and tree ensembles.
The authors tackled the problem of unreliable inference in hypothesis-free discovery from epidemiological data by proposing RuleSHAP, a framework that combines sparse Bayesian regression, tree ensembles, and Shapley values to detect and test complex patterns at the individual level, demonstrating its validity on simulated data and applying it to detect nonlinear interaction effects for high cholesterol and blood pressure in an epidemiological cohort.
In epidemiological settings, Machine Learning (ML) is gaining popularity for hypothesis-free discovery of risk (or protective) factors. Although ML is strong at discovering non-linearities and interactions, this power is currently compromised by a lack of reliable inference. Although local measures of feature effect can be combined with tree ensembles, uncertainty quantifications for these measures remain only partially available and oftentimes unsatisfactory. We propose RuleSHAP, a framework for using rule-based, hypothesis-free discovery that combines sparse Bayesian regression, tree ensembles and Shapley values in a one-step procedure that both detects and tests complex patterns at the individual level. To ease computation, we derive a formula that computes marginal Shapley values more efficiently for our setting. We demonstrate the validity of our framework on simulated data. To illustrate, we apply our machinery to data from an epidemiological cohort to detect and infer several effects for high cholesterol and blood pressure, such as nonlinear interaction effects between features like age, sex, ethnicity, BMI and glucose level.