How to Get Started in Machine Learning

A quick guide listing a few ways to get started with ML

Take a class at your university

Most 4-year universities have at least one course about machine learning. If your particular university doesn’t have a machine learning course, or if you’re not currently in university but you still wish to learn in a traditional way, you can buy a textbook. Some popular textbooks used by universities are Pattern Recognition and Machine Learning by Christopher Bishop, and Machine Learning by Kevin Murphy, and both do a fairly good job explaining the main concepts of machine learning. Though these texts are great at explaining machine learning concepts, it is worth noting that a good understanding of basic linear algebra is recommended.

Take an online course

There are many different online courses on machine learning, with the majority being free. A course I have personally completed is Machine Learning taught by Andrew Ng, a chief scientist at Baidu. This course taught by Andrew Ng is the most popular online introductory course on machine learning and I personally think it is the best source from introducing the topic to people with very little math experience. Also, there are many introductory machine learning courses, the Andrew Ng course is just the most popular and most comprehensive.

Read news outlets

While learning about a topic as complex as machine learning through news articles probably isn’t the best idea, there are news outlets that can help you demystify the field machine learning. Hacker News is a good news site for learning about what is currently happening in the world of machine learning. My only caveat to this recommendation is that Hacker News has a lot of material NOT related to machine learning, and some of the machine learning posts require a lot of prerequisite knowledge to understand. A blog I would personally recommend (Machine Learning is Fun) does a very good job of explaining machine learning in simple terms that don’t require a lot of prior knowledge.

Let me know if any of you started your machine learning journey down a different path!