Near optimal sample complexity for matrix and tensor normal models via geodesic convexity

Open in new window