How does momentum benefit deep neural networks architecture design? A few case studies (Q2079522): Difference between revisions

@@ Property / cites work @@
+THE HEAVY BALL WITH FRICTION METHOD, I. THE CONTINUOUS DYNAMICAL SYSTEM: GLOBAL EXPLORATION OF THE LOCAL MINIMA OF A REAL-VALUED FUNCTION BY ASYMPTOTIC ANALYSIS OF A DISSIPATIVE DYNAMICAL SYSTEM
+Normal rank
@@ Property / cites work @@
+Stable architectures for deep neural networks
@@ Property / cites work: Stable architectures for deep neural networks / rank @@
+Normal rank
@@ Property / cites work @@
+Q3320132
@@ Property / cites work: Q3320132 / rank @@
+Normal rank
@@ Property / cites work @@
+Some methods of speeding up the convergence of iteration methods
+Normal rank
@@ Property / cites work @@
+Adversarial defense via the data-dependent activation, total variation minimization, and adversarial training
+Normal rank
@@ Property / cites work @@
+Scheduled Restart Momentum for Accelerated Stochastic Gradient Descent
+Normal rank
@@ Property / cites work @@
+Graph interpolating activation improves both natural and robust accuracies in data-efficient deep learning
+Normal rank
@@ Property / cites work @@
+A variational perspective on accelerated methods in optimization
+Normal rank