Deep Expander Networks: Efficient Deep Networks from Graph Theory

نویسندگان

  • Ameya Prabhu
  • Girish Varma
  • Anoop M. Namboodiri
چکیده

Deep Neural Networks, while being unreasonably effective for several vision tasks, have their usage limited by the computational and memory requirements, both during training and inference stages. Analyzing and improving the connectivity patterns between layers of a network has resulted in several compact architectures like GoogleNet, ResNet and DenseNet-BC. In this work, we utilize results from graph theory to develop an efficient connection pattern between consecutive layers. Specifically, we use expander graphs that have excellent connectivity properties to develop a sparse network architecture, the deep expander network (X-Net). The X-Nets are shown to have high connectivity for a given level of sparsity. We also develop highly efficient training and inference algorithms for such networks. Experimental results show that we can achieve the similar or better accuracy as DenseNet-BC with twothirds the number of parameters and FLOPs on several image classification benchmarks. We hope that this work motivates other approaches to utilize results from graph theory to develop efficient network architectures.

برای دانلود رایگان 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

Detecting Overlapping Communities in Social Networks using Deep Learning

In network analysis, a community is typically considered of as a group of nodes with a great density of edges among themselves and a low density of edges relative to other network parts. Detecting a community structure is important in any network analysis task, especially for revealing patterns between specified nodes. There is a variety of approaches presented in the literature for overlapping...

متن کامل

The Diagnosis of Brucellosis in Rafsanjan City Using Deep Auto-Encoder Neural Networks

Introduction: Brucellosis is considered as one of the most important common infectious diseases between humans and animals. Considering the endemic nature of brucellosis and the existence of numerous reports of human and animal cases of brucellosis in Iran, the incidence of human brucellosis in Rafsanjan city was determined in the last 3 years (2016–2018). The main objective of this study was t...

متن کامل

VULNERABILITY ASSESSMENT OF WATER DISTRIBUTION NETWORKS: GRAPH THEORY METHOD

The main functional purpose of a water distribution network is to transport water from a source to several domestic and industrial units while at the same time satisfying various requirements on hydraulic response. All the water distribution networks perform two basic operations: firstly the water network needs to deliver adequate amounts of water to meet specific requirements, and secondly the...

متن کامل

Estimation of Hand Skeletal Postures by Using Deep Convolutional Neural Networks

Hand posture estimation attracts researchers because of its many applications. Hand posture recognition systems simulate the hand postures by using mathematical algorithms. Convolutional neural networks have provided the best results in the hand posture recognition so far. In this paper, we propose a new method to estimate the hand skeletal posture by using deep convolutional neural networks. T...

متن کامل

Simulate Congestion Prediction in a Wireless Network Using the LSTM Deep Learning Model

Achieved wireless networks since its beginning the prevalent wide due to the increasing wireless devices represented by smart phones and laptop, and the proliferation of networks coincides with the high speed and ease of use of the Internet and enjoy the delivery of various data such as video clips and games. Here's the show the congestion problem arises and represent   aim of the research is t...

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

عنوان ژورنال:
  • CoRR

دوره abs/1711.08757  شماره 

صفحات  -

تاریخ انتشار 2017