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

Link to Paper:

“EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks” - 2019

arxiv.org

Table of Contents

1. Introduction

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

Is there a principled method to scale up ConvNets that can achieve better accuracy and efficiency?

In previous work, it is common to scale only one of the three dimensions – depth, width, and image size. Though it is possible to scale two or three dimensions arbitrarily, arbitrary scaling requires tedious manual tuning and still often yields sub-optimal accuracy and efficiency.

image.png

<aside> <img src="/icons/report_green.svg" alt="/icons/report_green.svg" width="40px" />

ANSWER: Effective compound scaling method !

</aside>

Contributions