How To Kickstart Your Machine Learning Journey: 101 Machine Learning Guides Here

By Kimberly Cook | Dec 6, 2018 | 4326 Views

When it comes to pursuing a career in machine learning, it is not as simple as learning and getting ahead in their careers. Beginners are often bewildered with countless learning resources out there. This guide tells exactly where to kickstart your proficiency in Machine Learning.

Step 1: Math And Statistics
The first thing is the skill to understanding problems through a mathematical intuition. We would highly recommend starting with basics in linear algebra, then gradually moving to calculus. It may be hard to master them initially but given the time and practice with working, these areas will be familiar and comfortable to work on. Another subject that follows math is statistics. For any ML algorithm to be understood clearly, statistics is essential. Hence, fundamental knowledge in stats should be learned hand in hand.

By now, you might have easily thought of online websites such as Coursera or Udemy. Of course, these are good resources to learn at your own pace. The only hindrance is setting and following self-paced goals.

On the other hand, here are a few resources that are equally good at teaching the basics of math and statistics.

  • Linear Algebra tutorials by Kardi Teknomo - This interactive tutorial is a gem of a resource. Dr. Kardi Teknomo of Ateneo de Manila University explains every concept in a simple, easy to read the language.

  • Linear Algebra in Twenty Five Lectures - A concise course by the University of California, Davis scholars Tom Denton and Andrew Waldron.

  • Paul's Online Notes - A complete online resource (free to download) for math by Paul Dawkins of Lamar University. This resource is mainly centered around algebra and calculus.

  • Stat Trek - Online website for statistics.

  • Khan Academy - An all-time favorite among students, you would find a plethora of content on various areas of mathematics and statistics.

Step 2. Programming
For beginners, programming is sometimes dreadful to learn. Naturally, it might be intimidating and difficult at first. But with regular practice, this skill can be aced eventually. Coming to programming languages in ML, Python stands out top due to its versatility and ease in coding. Other languages such as R, Julia, and Java also help in building ML projects. Ultimately, programming should complement your math and stat knowledge.

Once mastered, try working on various problems and build small projects around ML. See how it solves a problem particularly by making use of math and statistics concepts. Sites like Kaggle and DataCamp are extremely good at testing codes and collaborating with peers and developers. In addition, forums like Stack Overflow are excellent to discuss problems and queries related to programming.   

These are the top resources that we found for learning Python, R and sites that cover programming in general.

  • - One of the best online resources for learning Python out there. The programmer behind this website, Harrison Kinsley (popularly known as Sentdex in the Python community) explains every aspect of Python perfectly!

  • Automate The Boring Stuff by Al Sweigart - Another very good online resource on Python. Programming is deconstructed right from scratch. In fact, the beauty lies in how simple tasks can be automated through Python.

  • ListenData for R - This website offers R tutorials for free. Extensive coverage of R concepts is what makes this site a catch for ML beginners preferring R over Python.

  • Code Project - A popular discussion forum exclusively for discussing programming queries in general. With a big developer user base, ML beginners can share their code, ask where they face problems in the code and work around ideas.

Step 3. Application-Oriented Approach
Learning is only fruitful if it is applied sensibly. Many people do the mistake of learning every algorithm in ML and forget where it actually helps in solving a problem. For beginners, it is suggested that they learn the popular and standard algorithms. A complicated algorithm is not always the solution for complex applications. It is all about how an ML problem is solved optimally.

Here are a few blogs which brilliantly tells about applications in ML, which is suggested for every beginner to go through them for realizing how ML is actually helpful.

  • Edwin Chen's blog - it explores requisite concepts of ML such as neural networks, deep learning etc. as well as the math behind it.  

Source: HOB