loss-based sampling

This research paper focuses on the convergence rates of loss and uncertainty-based active learning algorithms. The authors first provide conditions for a convergence rate guarantee and apply this to linear…