Graphical Perception of Saliency-based Model Explanations

TL;DR

This paper investigates how visualization design affects human perception of saliency-based explanations for deep learning models in visual recognition, revealing key factors that influence understanding.

Contribution

It introduces an experimental framework to study graphical perception of saliency explanations and identifies design factors impacting perception.

Findings

Visualization design significantly influences perception accuracy.

Type of alignment affects how well explanations are understood.

Saliency map qualities impact human perception.

Abstract

In recent years, considerable work has been devoted to explaining predictive, deep learning-based models, and in turn how to evaluate explanations. An important class of evaluation methods are ones that are human-centered, which typically require the communication of explanations through visualizations. And while visualization plays a critical role in perceiving and understanding model explanations, how visualization design impacts human perception of explanations remains poorly understood. In this work, we study the graphical perception of model explanations, specifically, saliency-based explanations for visual recognition models. We propose an experimental design to investigate how human perception is influenced by visualization design, wherein we study the task of alignment assessment, or whether a saliency map aligns with an object in an image. Our findings show that factors related to visualization design decisions, the type of alignment, and qualities of the saliency map all play important roles in how humans perceive saliency-based visual explanations.