Understanding Deep Neural Networks with Rectified Linear Units

Anirbit Mukherjee; Amitabh Basu; Raman Arora; Poorya Mianjy

Understanding Deep Neural Networks with Rectified Linear Units

Anirbit Mukherjee, Amitabh Basu, Raman Arora, Poorya Mianjy

Machine Learning and Robotics

Johns Hopkins University

Research output: Contribution to conference › Paper › peer-review

Original language	English
Publication status	Published - 2018
Event	International Conference on Learning Representations - Vancouver, Vancouver, Canada Duration: 30 Apr 2018 → 3 May 2018 https://iclr.cc/Conferences/2018

Conference

Conference	International Conference on Learning Representations
Abbreviated title	ICLR
Country/Territory	Canada
City	Vancouver
Period	30/04/18 → 3/05/18
Internet address	https://iclr.cc/Conferences/2018

Keywords

linear programming
polytopes
neural nets
complexity theory

Access to Document

https://openreview.net/forum?id=B1J_rgWRW

2 Article
1 Doctoral Thesis

Regularized Gradient Clipping Provably Trains Wide and Deep Neural Networks
Tucat, M., Mukherjee, A., Sun, M., Sen, P. & Rivasplata, O., 2025, In: Transactions on Machine Learning Research. 2025-June
Research output: Contribution to journal › Article › peer-review

Open Access
Global Convergence of SGD For Logistic Loss on Two Layer Neural Nets
Gopalani, P., Jha, S. & Mukherjee, A., 25 Feb 2024, In: Transactions on Machine Learning Research . 14, 1, p. 1-21 21 p.
Research output: Contribution to journal › Article › peer-review

Open Access
File
34 Downloads (Pure)
A Study of the Mathematics of Deep Learning
Mukherjee, A., 2020, (Accepted/In press) Johns Hopkins University.
Research output: Thesis › Doctoral Thesis

File

Cite this

@conference{d5dc33d5451948b481f0c4e767a1d78f,

title = "Understanding Deep Neural Networks with Rectified Linear Units",

keywords = "linear programming, polytopes, neural nets, complexity theory",

author = "Anirbit Mukherjee and Amitabh Basu and Raman Arora and Poorya Mianjy",

year = "2018",

language = "English",

note = "International Conference on Learning Representations , ICLR ; Conference date: 30-04-2018 Through 03-05-2018",

url = "https://iclr.cc/Conferences/2018",

}

TY - CONF

T1 - Understanding Deep Neural Networks with Rectified Linear Units

AU - Mukherjee, Anirbit

AU - Basu, Amitabh

AU - Arora, Raman

AU - Mianjy, Poorya

PY - 2018

Y1 - 2018

KW - linear programming

KW - polytopes

KW - neural nets

KW - complexity theory

M3 - Paper

T2 - International Conference on Learning Representations

Y2 - 30 April 2018 through 3 May 2018

ER -

Understanding Deep Neural Networks with Rectified Linear Units

Conference

Keywords

Access to Document

Research output

Regularized Gradient Clipping Provably Trains Wide and Deep Neural Networks

Global Convergence of SGD For Logistic Loss on Two Layer Neural Nets

A Study of the Mathematics of Deep Learning

Cite this