Can I read Manipulating and Measuring Model Interpretability on EtoBox?

Manipulating and Measuring Model Interpretability by Forough Poursabzi-Sangdeh; Daniel G Goldstein; Jake M Hofman; Jennifer Wortman Wortman Vaughan; Hanna Wallach is a scholarly article available to read on EtoBox.

What is Manipulating and Measuring Model Interpretability about?

With machine learning models being increasingly used to aid decision making even in high-stakes domains, there has been a growing interest in developing interpretable models. Although many supposedly interpretable models have been proposed, there have been relatively few experimental studies investigating whether these models achieve their intended efects, such as making people more closely follow a model's predictions when it is benefcial for them to do so or enabling them to detect when a model has made a mistake. We present a sequence of pre-registered experiments (N = 3, 800) in which we showed participants functionally identical models that varied only in two factors commonly thought to make machine learning models more or less interpretable: the number of features and the transparency of the model (i.e., whether the model internals are clear or black box). Predictably, participants who saw a clear model with few features could better simulate the model's predictions. However, we did not fnd that participants more closely followed its predictions. Furthermore, showing participants a clear model meant that they were less able to detect and correct for the model's sizable mistak

Author

Forough Poursabzi-Sangdeh; Daniel G Goldstein; Jake M Hofman; Jennifer Wortman Wortman Vaughan; Hanna Wallach

Publisher

ACM

Published

2021