Deep Convolutional Neural Networks

Pablo Ruiz
Towards Data Science
1 min readOct 11, 2018

The goal of this post is to serve as a nice introduction to deep architectures before diving to read the original publications where they are described.

I feel there is a lack of help in the research community. A little bit of the time of one researcher by making nice visualizations, dashboards, demos or even videos could save the time of all the researchers coming after him/her, and innovation will grow faster.

My contribution is by giving intuition in understanding the evolution of the so used deep convolutional neural networks as the default option for computer vision problems.

DenseNet — Example of how networks will be demystified

INDEX

0.1: Convolution Operations

0.2: 1x1 Convolution

(1): LeNet — LeCun 1998 — Paper ………………………….(TBI)

(2) AlexNet — Krizhevsky 2012 — Paper

(3): GoogLeNet / Inception — Szegedy 2014 — Paper…….(TBI)

(4): VGG — Simonyan / Zisserman 2014 — Paper…………(TBI)

(5): ResNets for ImageNet — He 2015 — Paper

(6): DenseNets for ImageNet

(7): FractalNets — Larsson — 2016 — Paper………………..(TBI)

(8): SENets — Hu — 2018 — Paper

(9): MobileNets — Howard — 2016 — Paper……………….(TBI)

--

--

Towards Data Science
Towards Data Science

Published in Towards Data Science

Your home for data science and AI. The world’s leading publication for data science, data analytics, data engineering, machine learning, and artificial intelligence professionals.

Pablo Ruiz
Pablo Ruiz

Written by Pablo Ruiz

Machine Learning @ Apple & DL Research Collaborator @ Harvard

Responses (4)