Towards a Unified Framework for Uncertainty-aware Nonlinear Variable Selection with Theoretical Guarantees

Wenying Deng · Beau Coker · Rajarshi Mukherjee · Jeremiah Liu · Brent Coull

Hall J #941

Keywords: [ variable selection ] [ Bayesian nonparametric ] [ posterior distribution ]

[ Abstract ]
[ Paper [ Poster [ OpenReview
Wed 30 Nov 2 p.m. PST — 4 p.m. PST

Abstract: We develop a simple and unified framework for nonlinear variable importance estimation that incorporates uncertainty in the prediction function and is compatible with a wide range of machine learning models (e.g., tree ensembles, kernel methods, neural networks, etc). In particular, for a learned nonlinear model $f(\mathbf{x})$, we consider quantifying the importance of an input variable $\mathbf{x}^j$ using the integrated partial derivative $\Psi_j = \Vert \frac{\partial}{\partial \mathbf{x}^j} f(\mathbf{x})\Vert^2_{P_\mathcal{X}}$. We then (1) provide a principled approach for quantifying uncertainty in variable importance by deriving its posterior distribution, and (2) show that the approach is generalizable even to non-differentiable models such as tree ensembles. Rigorous Bayesian nonparametric theorems are derived to guarantee the posterior consistency and asymptotic uncertainty of the proposed approach. Extensive simulations and experiments on healthcare benchmark datasets confirm that the proposed algorithm outperforms existing classical and recent variable selection methods.

Chat is not available.