$\color{black}\rule{365px}{3px}$

Link to Paper:

“Densely Connected Convolutional Networks” - 2017

arxiv.org

Table of Contents

1. Introduction

$\color{black}\rule{365px}{3px}$

Gao Huang | Zhuang Liu | Laurens Van der Maaten

Motivation

“Convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output”

Contributions

• DenseNet has several compelling advantages:
  • Alleviate the vanishing-gradient problem
  • Strengthen feature propagation
  • Encourage feature reuse
  • Substantially reduce the number of parameters(Counter-intuitive)