How to Learn Artificial Intelligence
@ danieljdick | Sunday, Feb 14, 2021 | 6 minutes read | Update at Sunday, Feb 14, 2021

I took the picture above while Eileen and I were near the backside of Half Dome in Yosemite traveling along Tioga Pass toward Mamouth soon before our son, David was born in 2012.

    Page: /

How Do You Learn Machine Learning?

Here is a list that may help you get into Artificial Intelligence and Machine Learning fast and effectively while avoiding burnout.

  1. Dig in! Have fun. Play with it.
  2. Explore the amazing free videos on YouTube.
  3. Build your hunger for it.
  4. Get excited about what it will do.
  5. See the magic behind it.
  6. Get into Fast.AI’s free courses and learn one easy way.
  7. When you’re ready, dig into a rigorous, deep course from Coursera’s Andrew Ng, or Geoffrey Hinton, or Udacity.
  8. If you get stuck, consider some Udemy courses to get a different perspective, but don’t stay stuck long.
  9. Go for perfect scores. You really do learn more when you push for perfect scores rather than passing scores.
  10. Play around with Kaggle competitions and free courses there.
  11. Spend time mastering data wrangling, cleaning, and pulling data from all kinds of sources.
  12. Learn engineering deployment of AI models into complete software engineering solutions.

Here’s a fun illustration somebody produced to show how different types of Machine Learning flow together.

Types of Machine Learning

Types of Machine Learning

This is taken from the website FavouriteBlog.com which provides excellent information engineers need to know.

So, what did I do wrong?

I approached it the hard way. I should have had more fun. Why? Fun drives happiness and success.

Google machine learning and artificial intelligence and you will find informal tutorials in machine learning. Easy tutorials. Fun tutorials. Look up teachers like Angela Yu. Check out YouTube videos and check out some Udemy courses when they’re cheap and on sale.

But one course you may want to check out is Fast.AI.

The main thing is not to get bogged down in theory and rigor until you know what that theory or rigor is for. Wait until you can feel the curiosity that will drive your success.

I’m a Curious Geek who Needs Hard Challenges

I’m a very, very curious person by nature. Ok. Yeah. Don’t laugh. I am a little weird. Ok. Don’t laugh again, please. I’m more than a little weird. I’m a math and computer and music geek. As a kid I had glasses held together with a band-aid, highwater plaid pants, pens in my shirt pocket, a slide rule, a fully adjustable camera, and I walked leaning forward at about a 10 degree angle. I wasn’t trying to look the part. The part was trying to look like me.

I played trombone. I rode a unicycle. I won contests in photography, math, and music, and was a 4-H All Star. If it seemed impossible, I had to figure out how to do it perfectly.

Should I Start With Fun or Rigor?

When Stanford Professor Andrew Ng and Toronto Professor Geoffrey Hinton offered online courses on Coursera, I signed up and pushed myself to obtain perfect scores in each course. And the courses were rigorous.

While these courses helped me understand theory well, they did not help me have fun or do much practically speaking. But when I got bored, burned out or frustrated, I would take a break, watch some YouTube videos and get another perspective, or I would sign up for a Udemy course on sale.

In the end, I feel the best thing you can do is have fun first, and then the theory will become the fun thing to do later. By doing that, you will become really good at doing practical work and you will also master the rigor when you are better positioned to enjoy and retain what you learn.

Have fun, and learn well. Who knows? You may find employers hunting you down and making offers you cannot refuse if you have a great portfolio of your work!

Some Videos and Courses to Check Out

The Age of AI is highly interesting and entertaining for opening anybody’s mind to what AI is all about.

Lex Fridman is perhaps one of the best known AI journalists out of M.I.T. interviewing the leaders in Artificial Intelligence, Science, Mathematics, and the Autonomous Vehicle Industry.

Mandy from Deep Lizard can help you get up and running with your own Machine Learning environment and teach you the practical basics quickly.

Lex Fridman interviews Andrew Ng who is the professor who taught me most of what I have learned. Perhaps he is the most widely known and respected Artificial Intelligence professors in the world.

Fast.AI is perhaps one of the best on-line training to dive in and learn practical skills fast and cheaply. It may be the very best place to get started.

Kaggle is an educational competition and fun education site that is highly active. Furthermore, it is a way to make money winning competitions if you are good. It is also a great way to become part of the Kaggle community. One thing to keep in mind is that while you will learn much about building AI models, there are other skills to develop. For instance, a big part of your work will involve acquiring the data, converting it into the proper form for a machine learning model, cleaning the data dealing with missing or buggy data, and deployment.

Geoffrey Hinton is very dry and rigorous to near perfection in his presentation. His course is said by many to be challenging to people with PhDs. Yet I find him to be one of my favorite professors teaching materials you will not easily find elsewhere. He was one of the three Turing Award winners this year and was one of those to promote using backpropagation for multi-layer neural networks for Deep Learning.

How to learn Machine Learning for free and plot your course to getting your first job in M.L. by Krish Naik.

Another quick-start into learning Machine Learning the easy way on your own by Tina Huang.

Copyright 2021 Dan Dick; all rights reserved