Optimal convex $M$-estimation via score matching

Open in new window