Residual Networks of Residual Networks: Multilevel Residual Networks

نویسندگان

  • Ke Zhang
  • Miao Sun
  • Tony X. Han
  • Xingfang Yuan
  • Liru Guo
  • Tao Liu
چکیده

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-networklike 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.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Relationship between Turbidity and Residual Chlorine and Microbial Quality of Drinking Water

Abstract Background and Objective: Safe drinking water is essential for health and health promotion is dependent on providing safe water. We aimed to determine the relationship between turbidity & residual chlorine and microbial quality of drinking water in Agh ghala. Material and Methods: In this descriptive-analytical study, 2079 water samples were collected from water networks of 78 ...

متن کامل

Analysis of the microbial quality in drinking water distribution networks using the logistic regression model in Dasht-e Azadegan county, an arid region in the southwest of Iran

The microbial quality of water plays a key role in community health. The present study aimed to determine the microbial quality of the drinking water distribution networks in the urban and rural areas of Dasht-e Azadegan County, Iran and assess the influential factors in the quality of drinking water.In this descriptive-analytical study, 907 drinking water samples were collected from the urban ...

متن کامل

The Least Cost Design of Water Distribution Networks Using Water Quality Constraints

Water quality is a very important issue related to people health, but it is not usually considered in water distribution networks (WDNs) design. In this paper, new constraints such as free residual chlorine and the quality performance index are incorporated into the least cost design of water distribution networks. EPANET2 was applied for the hydraulic and quality analysis of water distributio...

متن کامل

Modelling of Conventional and Severe Shot Peening Influence on Properties of High Carbon Steel via Artificial Neural Network

Shot peening (SP), as one of the severe plastic deformation (SPD) methods is employed for surface modification of the engineering components by improving the metallurgical and mechanical properties. Furthermore artificial neural network (ANN) has been widely used in different science and engineering problems for predicting and optimizing in the last decade. In the present study, effects of conv...

متن کامل

Using a Fuzzy Rule-based Algorithm to Improve Routing in MPLS Networks

Today, the use of wireless and intelligent networks are widely used in many fields such as information technology and networking. There are several types of these networks that MPLS networks are one of these types. However, in MPLS networks there are issues and problems in the design and implementation discussion, for example security, throughput, losses, power consumption and so on. Basically,...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • CoRR

دوره abs/1608.02908  شماره 

صفحات  -

تاریخ انتشار 2016