On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length.
Stanislaw Jastrzebski, Zachary Kenton, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos J. Storkey
Browse the full ICLR paper archive.
Stanislaw Jastrzebski, Zachary Kenton, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos J. Storkey
Browse the full ICLR paper archive.