Skip to yearly menu bar Skip to main content


(4 events)   Time zone: Conference local time zone  
Show all
Toggle Poster Visibility
Oral
Wed Jul 11 04:00 PM -- 04:20 PM (CEST) @ K1 + K2
Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networks
In Deep Learning (Theory) 1
Lechao Xiao · Yasaman Bahri · Jascha Sohl-Dickstein · Samuel Schoenholz · Jeffrey Pennington
[ PDF]  [ Video] 
Oral
Wed Jul 11 04:20 PM -- 04:40 PM (CEST) @ K1 + K2
The Dynamics of Learning: A Random Matrix Approach
In Deep Learning (Theory) 1
Zhenyu Liao · Romain Couillet
[ PDF]  [ Video] 
Oral
Wed Jul 11 04:40 PM -- 04:50 PM (CEST) @ K1 + K2
On the Optimization of Deep Networks: Implicit Acceleration by Overparameterization
In Deep Learning (Theory) 1
Sanjeev Arora · Nadav Cohen · Elad Hazan
[ PDF]  [ Video] 
Oral
Wed Jul 11 04:50 PM -- 05:00 PM (CEST) @ K1 + K2
Deep Linear Networks with Arbitrary Loss: All Local Minima Are Global
In Deep Learning (Theory) 1
Thomas Laurent · James von Brecht
[ PDF]  [ Video]