Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Residual Networks of Residual Networks: Multilevel Residual Networks

Abstract: 

A residual-networks family with hundreds or even thousands of layers dominates major image recognition tasks, but building a network by simply stacking residual blocks inevitably limits its optimization ability. This paper proposes a novel residual-network architecture, Residual networks of Residual networks (RoR), to dig the optimization ability of residual networks. RoR substitutes optimizing residual mapping of residual mapping for optimizing original residual mapping. In particular, RoR adds level-wise shortcut connections upon original residual networks to promote the learning capability of residual networks. More importantly, RoR can be applied to various kinds of residual networks (ResNets, Pre-ResNets and WRN) and significantly boost their performance. Our experiments demonstrate the effectiveness and versatility of RoR, where it achieves the best performance in all residual-network-like structures. Our RoR-3-WRN58-4+SD models achieve new state-of-the-art results on CIFAR-10, CIFAR-100 and SVHN, with test errors 3.77%, 19.73% and 1.59%, respectively. RoR-3 models also achieve state-of-the-art results compared to ResNets on ImageNet data set.

up
0 users have voted:

Paper Details

Authors:
Ke Zhang, Miao Sun, Tony Han, Xingfang Yuan, Liru Guo, Tao Liu
Submitted On:
12 September 2017 - 10:28pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Ke Zhang
Paper Code:
4650
Document Year:
2017
Cite

Document Files

ICIP 2017.pdf

(8 downloads)

Subscribe

[1] Ke Zhang, Miao Sun, Tony Han, Xingfang Yuan, Liru Guo, Tao Liu, "Residual Networks of Residual Networks: Multilevel Residual Networks", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1954. Accessed: Sep. 24, 2017.
@article{1954-17,
url = {http://sigport.org/1954},
author = {Ke Zhang; Miao Sun; Tony Han; Xingfang Yuan; Liru Guo; Tao Liu },
publisher = {IEEE SigPort},
title = {Residual Networks of Residual Networks: Multilevel Residual Networks},
year = {2017} }
TY - EJOUR
T1 - Residual Networks of Residual Networks: Multilevel Residual Networks
AU - Ke Zhang; Miao Sun; Tony Han; Xingfang Yuan; Liru Guo; Tao Liu
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1954
ER -
Ke Zhang, Miao Sun, Tony Han, Xingfang Yuan, Liru Guo, Tao Liu. (2017). Residual Networks of Residual Networks: Multilevel Residual Networks. IEEE SigPort. http://sigport.org/1954
Ke Zhang, Miao Sun, Tony Han, Xingfang Yuan, Liru Guo, Tao Liu, 2017. Residual Networks of Residual Networks: Multilevel Residual Networks. Available at: http://sigport.org/1954.
Ke Zhang, Miao Sun, Tony Han, Xingfang Yuan, Liru Guo, Tao Liu. (2017). "Residual Networks of Residual Networks: Multilevel Residual Networks." Web.
1. Ke Zhang, Miao Sun, Tony Han, Xingfang Yuan, Liru Guo, Tao Liu. Residual Networks of Residual Networks: Multilevel Residual Networks [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1954