Sparsity-Aware Orthogonal Initialization of Deep Neural Networks-Reference-Cited by-同舟云学术

Sparsity-Aware Orthogonal Initialization of Deep Neural Networks

Author:

Esguerra Kiara¹^ORCID,Nasir Muneeb¹,Tang Tong Boon¹^ORCID,Tumian Afidalina²,Ho Eric Tatt Wei¹^ORCID

Affiliation:

1. Department of Electrical and Electronics Engineering, Universiti Teknologi PETRONAS, Seri Iskandar, Perak, Malaysia

2. Petroliam Nasional Berhad (PETRONAS), Kuala Lumpur, Malaysia

Funder

Yayasan Universiti Teknologi PETRONAS in collaboration with PETRONAS Research Sdn Bhd

APC was funded by the Petroleum Research Fund of Malaysia

PETRONAS Research Sdn Bhd and in collaboration with Universiti Teknologi PETRONAS

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

General Engineering,General Materials Science,General Computer Science,Electrical and Electronic Engineering

Link

Reference64 articles.

1. A signal propagation perspective for pruning neural networks at initialization;lee;Proc Int Conf Learn Represent,2019

2. Spectrally-normalized margin bounds for neural networks;bartlett;Proc Adv Neural Inf Process Syst,2017

3. Pruning neural networks without any data by iteratively conserving synaptic flow;tanaka;Proc Adv Neural Inf Process Syst,2020

4. Exact solutions to the nonlinear dynamics of learning in deep linear neural networks;saxe;Proc 2nd Int Conf Learn Represent (ICLR),2014

Cited by 1 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献