Deep Learning — Andrew Ng Coursera Specialization

Kevin Stock
4 min readAug 11, 2017

--

Andrew Ng — after stepping down as Baidu’s Chief Scientist — left much of the AI world wondering “what’s next?”

Prior to Baudu, Ng co-founded Coursera…and so it shouldn’t be a huge surprise that he is back…

…with his new course Deep Learning

Deep Learning Specialization

This new Coursera Specialization is broken into 5 different courses.

The objective of the Specialization is to learn the foundations of Deep Learning, including how to build neural networks, lead machine learning projects, and quite a bit more (like: convolutional networks, RNNs, LSTM, Adam, Dropout, BatchNorm, Xavier/He initialization).

5 Course List

  1. Neural Networks and Deep Learning

2. Improving Deep Neural Networks

3. Structuring Machine Learning Projects

4. Convolutional Neural Networks

5. Sequence Models

Projects Overview

Throughout the Specialization, students are asked to complete numerous projects such as deep learning models in healthcare, autonomous driving, sign language reading, music generation, and natural language processing.

Who is it for?

On the course registration it says:

However…

If you dig a little deeper, there are a number or prerequisites — not to enroll — but to ensure you can get value out of the course.

To get the most out of this Specialization you need to have at least intermediate level Python skills and experience with Machine Learning.

It’s probably not a bad idea to take his, outrageously popular, Coursera “Machine Learning” course first.

When does it start?

Coursera Specializations run in “Sessions.”

The 1st session of this course starts August 15th, and you generally have a few days after the start date in which you can still enroll.

If you miss the first session enrollment date, don’t sweat it. The sessions generally re-open each month.

If you are wondering what is the point of having “sessions” and why can’t you enroll whenever you want, the reason is to maintain forum/community continuity as well as having soft deadlines in batches.

If you miss a deadline, it’s not a big deal, you can just fall into the next incoming batch/session.

Siraj Raval’s Deep Learning Foundation (Udacity Nanodegree) vs Andrew Ng’s Deep Learning (Coursera Specialization)

Which to take?!

This is a tough question.

But…

I think I have an answer…

It’s does not have to be an either/or question — I think taking both is the way to go.

Just not at the same time :)

You will undoubtedly gain more understanding and will be presented new material in each.

One will enhance the other.

How much does Andrew Ng’s Deep Learning Specialization cost?

Coursera has adopted a subscription model instead of a one-time payment for their Specializations.

With this Specialization you get a 7 day free trial and then it’s $49/month (no continued free version).

Based on it taking an estimated ~4 months to complete, you’re looking at ~$200…though if you can do it faster you can save some money, or slower…and the tab just keeps running.

Trick:

Register through CredEd and get discounts/rewards ;)

Is it worth it? Do employers care about these credentials?

Deep Learning is highly in-demand and will continue to be highly in-demand for the foreseeable future.

These alternative credentials — whether it be a Coursera Specialization or a Udacity Nanodegree — are not only gaining acceptance among employers, I believe they are going to be the cornerstone of the “ePortfolio” of the future.

At CredEd it’s our mission to connect online learning and opportunity, so if you are wondering what kind of job you can get…head on over to CredEd and let us help.

Learn more: Literally. Us learning nerds hangout here.

➤ If you sign up on CredEd, make sure to shoot me a message!

➤ Do you like social media? If so, lets connect: Twitter, Instagram, Facebook…I like Snapchat too, LinkedIn not so much…

➤ Do you have a question? Ask me on Quora…I answer questions daily!

Kevin Stock is the CEO at CredEd, inventor of the NED (sleep apnea and snoring device), and author of Yourdrum (←read this. just do it.)

--

--