Representing smooth functions as compositions of near-identity functions
with implications for deep network optimization

v1v2 (latest)

Representing smooth functions as compositions of near-identity functions with implications for deep network optimization

13 April 2018

Peter L. Bartlett

ArXiv (abs)PDF HTML

Papers citing "Representing smooth functions as compositions of near-identity functions with implications for deep network optimization"

12 / 12 papers shown

Title
Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness Jeremiah Zhe Liu Zi Lin Shreyas Padhy Dustin Tran Tania Bedrax-Weiss Balaji Lakshminarayanan UQCV BDL 171 451 0 17 Jun 2020
Gradient descent with identity initialization efficiently learns positive definite linear transformations by deep residual networks Peter L. Bartlett D. Helmbold Philip M. Long 81 116 0 16 Feb 2018
The Reversible Residual Network: Backpropagation Without Storing Activations Aidan Gomez Mengye Ren R. Urtasun Roger C. Grosse 74 550 0 14 Jul 2017
Identity Matters in Deep Learning Moritz Hardt Tengyu Ma OOD 87 398 0 14 Nov 2016
Exponential expressivity in deep neural networks through transient chaos Ben Poole Subhaneil Lahiri M. Raghu Jascha Narain Sohl-Dickstein Surya Ganguli 90 592 0 16 Jun 2016
Faster Eigenvector Computation via Shift-and-Invert Preconditioning Dan Garber Laurent Dinh Chi Jin Jascha Narain Sohl-Dickstein Samy Bengio Praneeth Netrapalli Aaron Sidford 272 3,702 0 26 May 2016
Learning Functions: When Is Deep Better Than Shallow H. Mhaskar Q. Liao T. Poggio 69 144 0 03 Mar 2016
Representation Benefits of Deep Feedforward Networks Matus Telgarsky 76 242 0 27 Sep 2015
Gradient-based Hyperparameter Optimization through Reversible Learning D. Maclaurin David Duvenaud Ryan P. Adams DD 227 945 0 11 Feb 2015
Breaking the Curse of Dimensionality with Convex Neural Networks Francis R. Bach 184 706 0 30 Dec 2014
On the Number of Linear Regions of Deep Neural Networks Guido Montúfar Razvan Pascanu Kyunghyun Cho Yoshua Bengio 88 1,254 0 08 Feb 2014
Predicting Parameters in Deep Learning Misha Denil B. Shakibi Laurent Dinh MarcÁurelio Ranzato Nando de Freitas OOD 200 1,319 0 03 Jun 2013