Abstract
Through past experiences deploying what we call usable ML (one step beyondexplainable ML, including both explanations and other augmenting information)to real-world domains, we have learned three key lessons. First, manyorganizations are beginning to hire people who we call ``bridges'' because theybridge the gap between ML developers and domain experts, and these people filla valuable role in developing usable ML applications. Second, a configurablesystem that enables easily iterating on usable ML interfaces duringcollaborations with bridges is key. Finally, there is a need for continuous,in-deployment evaluations to quantify the real-world impact of usable ML.Throughout this paper, we apply these lessons to the task of wind turbinemonitoring, an essential task in the renewable energy domain. Turbine engineersand data analysts must decide whether to perform costly in-personinvestigations on turbines to prevent potential cases of brakepad failure, andwell-tuned usable ML interfaces can aid with this decision-making process.Through the applications of our lessons to this task, we hope to demonstratethe potential real-world impact of usable ML in the renewable energy domain.