Spring 2018 


On the Optimization of Deep Networks: Implicit Acceleration by Overparameterization 
Zhan Lin 
Mar 30 
An Alternative View: When Does SGD Escape Local Minima? 
Sen Na 
Mar 15 
Stronger generalization bounds for deep nets via a compression approach 
Xialiang Dou 
Apr 6 
Risk and parameter convergence of logistic regression 
Sen Na 
Apr 13, 20 
Gradient Descent Quantizes ReLU Network Features 
Zhan Lin 
May 25 
Generalization in Machine Learning via Analytical Learning Theory 


Sever: A Robust MetaAlgorithm for Stochastic Optimization 


Representing smooth functions as compositions of nearidentity functions with implications for deep network optimization 


Are ResNets Provably Better than Linear Predictors? 
Xialiang Dou 
Apr 27 
Representing smooth functions as compositions of nearidentity functions with implications for deep network optimization 
Zhan Lin 
May 4 
On the Convergence of Adam and Beyond 
Xialiang Dou 
May 11 