|
Poster
|
Wed 7:45
|
How to Start Training: The Effect of Initialization and Architecture
Boris Hanin · David Rolnick
|
|
|
Poster
|
Wed 14:00
|
Are ResNets Provably Better than Linear Predictors?
Ohad Shamir
|
|
|
Poster
|
Tue 14:00
|
Training Neural Networks Using Features Replay
Zhouyuan Huo · Bin Gu · Heng Huang
|
|
|
Poster
|
Tue 7:45
|
Bayesian Distributed Stochastic Gradient Descent
Michael Teng · Frank Wood
|
|
|
Poster
|
Thu 14:00
|
Constructing Fast Network through Deconstruction of Convolution
Yunho Jeon · Junmo Kim
|
|
|
Poster
|
Tue 14:00
|
Provably Correct Automatic Sub-Differentiation for Qualified Programs
Sham Kakade · Jason Lee
|
|
|
Poster
|
Tue 7:45
|
Training Deep Models Faster with Robust, Approximate Importance Sampling
Tyler Johnson · Carlos Guestrin
|
|
|
Poster
|
Wed 14:00
|
How SGD Selects the Global Minima in Over-parameterized Learning: A Dynamical Stability Perspective
Lei Wu · Chao Ma · Weinan E
|
|
|
Poster
|
Tue 7:45
|
Which Neural Net Architectures Give Rise to Exploding and Vanishing Gradients?
Boris Hanin
|
|
|
Poster
|
Thu 7:45
|
Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search
Zhuwen Li · Qifeng Chen · Vladlen Koltun
|
|
|
Poster
|
Wed 7:45
|
On the Local Hessian in Back-propagation
Huishuai Zhang · Wei Chen · Tie-Yan Liu
|
|
|
Poster
|
Tue 7:45
|
TETRIS: TilE-matching the TRemendous Irregular Sparsity
Yu Ji · Ling Liang · Lei Deng · Youyang Zhang · Youhui Zhang · Yuan Xie
|
|
