IMPROVING USABILITY OF MACHINE LEARNING MODEL EXPLANATIONS: EXPERIMENTS WITH VISUAL AND VERBAL COGNITIVE LOAD

Abstract

Research in eXplainable Artificial Intelligence (XAI) has seen a flurry of activity, resulting in many new interpretable machine learning models and explanation techniques. However, it is not clear how useful these explanations are for non-expert end users. In this thesis we investigate how to improve the usability of explanations through theory-driven integration of cognitive load. We draw from cognitive psychology theories in graph comprehension to model cognitive load as visual chunks in line and bar chart visualizations of Generalized Additive Models (GAMs) and propose Cognitive-GAM(COGAM) which helps select explanations for desired cognitive load and accuracy. Next, we present CaptionCOGAM which selects salient visual information from graphs of COGAM and verbalizes them with extensible templates to automatically generate captions that balance verbal cognitive load with human interpretability. COGAM and CaptionCOGAM together provide a unified framework to characterize the criteria of explanation parsimony, visual and verbal cognitive loads, and human interpretability.