Explainable Machine Learning in Deployment

Explainable machine learning offers the potential to provide stakeholders with insights into model behavior by using various methods such as feature importance scores, counterfactual explanations, or influential training data. Yet there is little understanding of how organizations use these methods in practice. This study explores how organizations view and use explainability for stakeholder consumption. We find that, currently, the majority of deployments are not for end users affected by the model but rather for machine learning engineers, who use explainability to debug the model itself. There is thus a gap between explainability in practice and the goal of transparency, since explanations primarily serve internal stakeholders rather than external ones. Our study synthesizes the limitations of current explainability techniques that hamper their use for end users. To facilitate end user interaction, we develop a framework for establishing clear goals for explainability. We end by discussing concerns raised regarding explainability.

Focus: Methods or Design
Source: FAT 2020
Readability: Expert
Type: Website Article
Open Source: No
Keywords: machine learning, explainability, transparency, deployed systems, qualitative study
Learn Tags: Design/Methods Framework AI and Machine Learning
Summary: A study that analyzes the limits of current explainable machine learning techniques for end users and proposes a framework for establishing clear guidelines for explainability.