How (In)Accurate is Machine Learning?

Chicago Booth’s Max Farrell, Tengyuan Liang, and Sanjog Misra have sought to quantify the uncertainty in machine learning analysis so that decision makers can take it into account while making decisions. Executives and others are increasingly using data when assessing business policies, comparing marketing strategies, and making other decisions. And to analyze that data, they use machine learning, the results of which are used to determine decisions and action. In the past few years, machine-learning methods have come to dominate data analysis in academia and industry. One type of learning in particular—deep learning, where computers learn through iterations to recognize important features—has become a mainstay in modern business practice. It is at the base of many applications, from digital image recognition to language processing and virtual assistants such as Apple’s Siri and the Amazon Alexa. But can an executive trust a recommendation generated solely by machine learning?