Timezone: »
Poster
Variable Importance Using Decision Trees
Jalil Kazemitabar · Arash Amini · Adam Bloniarz · Ameet S Talwalkar
Decision trees and random forests are well established models that not only offer good predictive performance, but also provide rich feature importance information. While practitioners often employ variable importance methods that rely on this impurity-based information, these methods remain poorly characterized from a theoretical perspective. We provide novel insights into the performance of these methods by deriving finite sample performance guarantees in a high-dimensional setting under various modeling assumptions. We further demonstrate the effectiveness of these impurity-based methods via an extensive set of simulations.
Author Information
Jalil Kazemitabar (University of California, Los Angeles)
Arash Amini (UCLA)
Adam Bloniarz (Google)
Ameet S Talwalkar (CMU)
More from the Same Authors
-
2021 : Simulated User Studies for Explanation Evaluation »
Valerie Chen · Gregory Plumb · Nicholay Topin · Ameet S Talwalkar -
2021 : Bayesian Persuasion for Algorithmic Recourse »
Keegan Harris · Valerie Chen · Joon Kim · Ameet S Talwalkar · Hoda Heidari · Steven Wu -
2021 : Bayesian Persuasion for Algorithmic Recourse »
Keegan Harris · Valerie Chen · Joon Kim · Ameet S Talwalkar · Hoda Heidari · Steven Wu -
2022 Poster: Target alignment in truncated kernel ridge regression »
Arash Amini · Richard Baumgartner · Dai Feng -
2021 : [S9] Simulated User Studies for Explanation Evaluation »
Valerie Chen · Gregory Plumb · Nicholay Topin · Ameet S Talwalkar -
2021 Poster: Label consistency in overfitted generalized $k$-means »
Linfan Zhang · Arash Amini -
2020 Poster: The Potts-Ising model for discrete multivariate data »
Zahra Razaee · Arash Amini -
2019 Poster: Globally optimal score-based learning of directed acyclic graphs in high-dimensions »
Bryon Aragam · Arash Amini · Qing Zhou -
2017 Poster: Federated Multi-Task Learning »
Virginia Smith · Chao-Kai Chiang · Maziar Sanjabi · Ameet S Talwalkar -
2016 : Invited Talk: Paleo: A Performance Model for Deep Neural Networks (Ameet Talwalkar, UCLA) »
Ameet S Talwalkar -
2016 Poster: Yggdrasil: An Optimized System for Training Deep Decision Trees at Scale »
Firas Abuzaid · Joseph K Bradley · Feynman Liang · Andrew Feng · Lee Yang · Matei Zaharia · Ameet S Talwalkar -
2014 Workshop: Distributed Machine Learning and Matrix Computations »
Reza Zadeh · Ion Stoica · Ameet S Talwalkar -
2013 Poster: Bayesian inference as iterated random functions with applications to sequential inference in graphical models »
Arash Amini · XuanLong Nguyen -
2013 Spotlight: Bayesian inference as iterated random functions with applications to sequential inference in graphical models »
Arash Amini · XuanLong Nguyen -
2011 Workshop: Sparse Representation and Low-rank Approximation »
Ameet S Talwalkar · Lester W Mackey · Mehryar Mohri · Michael W Mahoney · Francis Bach · Mike Davies · Remi Gribonval · Guillaume R Obozinski -
2011 Poster: Divide-and-Conquer Matrix Factorization »
Lester W Mackey · Ameet S Talwalkar · Michael Jordan -
2010 Workshop: Low-rank Methods for Large-scale Machine Learning »
Arthur Gretton · Michael W Mahoney · Mehryar Mohri · Ameet S Talwalkar -
2009 Poster: Ensemble Nystrom Method »
Sanjiv Kumar · Mehryar Mohri · Ameet S Talwalkar