Saved in:
Bibliographic Details
Main Authors: Piratla, Vihari, Heo, Juyeon, Collins, Katherine M., Singh, Sukriti, Weller, Adrian
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.08063
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913300405551104
author Piratla, Vihari
Heo, Juyeon
Collins, Katherine M.
Singh, Sukriti
Weller, Adrian
author_facet Piratla, Vihari
Heo, Juyeon
Collins, Katherine M.
Singh, Sukriti
Weller, Adrian
contents Model explanations can be valuable for interpreting and debugging predictive models. We study a specific kind called Concept Explanations, where the goal is to interpret a model using human-understandable concepts. Although popular for their easy interpretation, concept explanations are known to be noisy. We begin our work by identifying various sources of uncertainty in the estimation pipeline that lead to such noise. We then propose an uncertainty-aware Bayesian estimation method to address these issues, which readily improved the quality of explanations. We demonstrate with theoretical analysis and empirical evaluation that explanations computed by our method are robust to train-time choices while also being label-efficient. Further, our method proved capable of recovering relevant concepts amongst a bank of thousands, in an evaluation with real-datasets and off-the-shelf models, demonstrating its scalability. We believe the improved quality of uncertainty-aware concept explanations make them a strong candidate for more reliable model interpretation. We release our code at https://github.com/vps-anonconfs/uace.
format Preprint
id arxiv_https___arxiv_org_abs_2312_08063
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Estimation of Concept Explanations Should be Uncertainty Aware
Piratla, Vihari
Heo, Juyeon
Collins, Katherine M.
Singh, Sukriti
Weller, Adrian
Machine Learning
Artificial Intelligence
Computation and Language
Model explanations can be valuable for interpreting and debugging predictive models. We study a specific kind called Concept Explanations, where the goal is to interpret a model using human-understandable concepts. Although popular for their easy interpretation, concept explanations are known to be noisy. We begin our work by identifying various sources of uncertainty in the estimation pipeline that lead to such noise. We then propose an uncertainty-aware Bayesian estimation method to address these issues, which readily improved the quality of explanations. We demonstrate with theoretical analysis and empirical evaluation that explanations computed by our method are robust to train-time choices while also being label-efficient. Further, our method proved capable of recovering relevant concepts amongst a bank of thousands, in an evaluation with real-datasets and off-the-shelf models, demonstrating its scalability. We believe the improved quality of uncertainty-aware concept explanations make them a strong candidate for more reliable model interpretation. We release our code at https://github.com/vps-anonconfs/uace.
title Estimation of Concept Explanations Should be Uncertainty Aware
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2312.08063