Convergence and Implicit Bias of Gradient Descent on Continual Linear Classification

Open in new window