Gradient Descent with Identity Initialization Efficiently Learns Positive-Definite Linear Transformations by Deep Residual Networks
['Peter L. Bartlett', 'David P. Helmbold', 'Philip M. Long']
/
Neural Computation
/ Vol. 31
/ No. 3
0.0 / 5
0
Reviews
You must log in to post, edit and delete a review.
Log in
Earn 50 points when your review is published.
About KnowledgeRank Points
No reviews have been posted yet. Be the first to write one!