On characterizing optimal learning trajectories in a class of learning problems

Open in new window