Non-IID Learning

Abstract

Real-life AI systems are non-IID, i.e., their variables are unlikely independent and drawn from the same distribution. Instead, non-IIDness is a common characteristic and complexity of real-life systems, where variables, objects, and subsystems are coupled/interactive and heterogeneous. This issue highlights this important theme on Non-IID Learning with six feature articles. In addition, four columns highlight expert opinions on beyond i.i.d., trustworthy AI, data-driven predictive maintenance, and secrets for data science deployments, respectively.