SDCA without Duality, Regularization, and Individual Convexity
Autori:
Shai Shalev-Shwartz
Pubblicato in:
ICML, 2016
Editore:
ICML
Failures of Gradient-Based Deep Learning
Autori:
Shai Shalev-Shwartz, Ohad Shamir, Shaked Shammah
Pubblicato in:
ICML, 2017
Editore:
ICML
"Decoupling ""when to update"" from ""how to update"""
Autori:
Eran Malach, Shai Shalev-Shwartz
Pubblicato in:
NIPS, 2017
Editore:
Curran Associates, Inc.
Complexity theoretic limitations on learning DNF's
Autori:
Amit Daniely, Shai Shalev-Shwartz.
Pubblicato in:
COLT, 2016
Editore:
COLT
On Graduated Optimization for Stochastic Non-Convex Problems
Autori:
Elad Hazan, Kfir Y. Levy, Shai Shalev-Shwartz
Pubblicato in:
ICML, 2016
Editore:
ICML
Fast Rates for Empirical Risk Minimization of Strict Saddle Problems
Autori:
Alon Gonen, Shai Shalev-Shwartz
Pubblicato in:
COLT, 2017
Editore:
COLT
SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data
Autori:
Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz
Pubblicato in:
International Conference on Learning Representations (ICLR), 2018
Editore:
ICLR
Solving Ridge Regression using Sketched Preconditioned SVRG
Autori:
Alon Gonen, Francesco Orabona, Shai Shalev-Shwartz
Pubblicato in:
ICML, 2016
Editore:
ICML
On the Expressive Power of Overlapping Architectures of Deep Learning
Autori:
Or Sharir, Amnon Shashua
Pubblicato in:
ICLR, 2018
Editore:
ICLR
Sum-Product-Quotient Networks
Autori:
Or Sharir, Amnon Shashua
Pubblicato in:
AISTATS, 2018
Editore:
AISTATS
On the Long-Term Memory of Deep Recurrent Networks
Autori:
Yoav Levine, Or Sharir, Alon Ziv, Amnon Shashua
Pubblicato in:
ICLR, 2018
Editore:
ICLR workshop
